Severity Distribution Modeling Action Set

={{

format="string",

formattedLength=integer,

label="string",

name="variable-name",

nfd=integer,

nfl=integer

}, {...}}

where="where-expression"

}

caslib="string",

dataSourceOptions={key-1=any-list-or-data-type-1 <, key-2=any-list-or-data-type-2, ...>},

groupBy

={{

format="string",

formattedLength=integer,

label="string",

name="variable-name",

nfd=integer,

nfl=integer

}, {...}},

groupByMode="NOSORT" | "REDISTRIBUTE",

name="table-name",

where="where-expression",

casLib="string"

name="table-name"

={{

format="string",

formattedLength=integer,

label="string",

name="variable-name",

nfd=integer,

nfl=integer

}, {...}}

where="where-expression"

}

={{

distribution="string",

parameters

={{

name="string",

value=double

}, {...}}

}, {...}},

={{

name="variable-name",

role="LC" | "LT" | "RC" | "RT" | "TARGET"

}, {...}},

={{

details=TRUE | FALSE,

name="string",

noEffect=TRUE | FALSE,

stdize=TRUE | FALSE,

vars={"variable-name-1" <, "variable-name-2", ...>},

weight={"variable-name-1" <, "variable-name-2", ...>}

}, {...}},

nClassLevelsPrint=integer,

noConstFitStats=TRUE | FALSE,

SASProgram="string",

symbol="string"

outDetailLevel=integer,

covout=TRUE | FALSE,

selectOut=TRUE | FALSE,

caslib="string"

compress=TRUE | FALSE

indexVars={"variable-name-1" <, "variable-name-2", ...>}

label="string"

lifetime=64-bit-integer

maxMemSize=64-bit-integer

memoryFormat="DVR" | "INHERIT" | "STANDARD"

name="table-name"

onDemand=TRUE | FALSE

promote=TRUE | FALSE

replace=TRUE | FALSE

replication=integer

threadBlockSize=64-bit-integer

timeStamp="string"

where={"string-1" <, "string-2", ...>}

zeroEstForNotInModel=TRUE | FALSE

caslib="string",

compress=TRUE | FALSE,

indexVars={"variable-name-1" <, "variable-name-2", ...>},

label="string",

lifetime=64-bit-integer,

maxMemSize=64-bit-integer,

memoryFormat="DVR" | "INHERIT" | "STANDARD",

name="table-name",

onDemand=TRUE | FALSE,

promote=TRUE | FALSE,

replace=TRUE | FALSE,

replication=integer,

threadBlockSize=64-bit-integer,

timeStamp="string",

where={"string-1" <, "string-2", ...>}

casOut

caslib="string"

compress=TRUE | FALSE

indexVars={"variable-name-1" <, "variable-name-2", ...>}

label="string"

lifetime=64-bit-integer

maxMemSize=64-bit-integer

memoryFormat="DVR" | "INHERIT" | "STANDARD"

name="table-name"

onDemand=TRUE | FALSE

promote=TRUE | FALSE

replace=TRUE | FALSE

replication=integer

threadBlockSize=64-bit-integer

timeStamp="string"

where={"string-1" <, "string-2", ...>}

copyVars="ALL" | "ALL_MODEL" | "ALL_NUMERIC" | {"variable-name-1" <, "variable-name-2", ...>},

quantiles

names={"string-1" <, "string-2", ...>}

nDecimal=integer

points

={{

p=double

}, {...}}

scoreFunctions

={{

arguments={double-1 <, double-2, ...>},

name="string",

variable="string"

}, {...}}

groupByVarsRaw=TRUE | FALSE,

includeAll=TRUE | FALSE,

names={"string-1" <, "string-2", ...>} | {key-1={casouttable-1} <, key-2={casouttable-2}, ...>},

repeated=TRUE | FALSE,

replace=TRUE | FALSE

commonPackage=TRUE | FALSE,

outById

caslib="string"

compress=TRUE | FALSE

indexVars={"variable-name-1" <, "variable-name-2", ...>}

label="string"

lifetime=64-bit-integer

maxMemSize=64-bit-integer

memoryFormat="DVR" | "INHERIT" | "STANDARD"

name="table-name"

onDemand=TRUE | FALSE

promote=TRUE | FALSE

replace=TRUE | FALSE

replication=integer

threadBlockSize=64-bit-integer

timeStamp="string"

where={"string-1" <, "string-2", ...>}

}

caslib="string",

compress=TRUE | FALSE,

indexVars={"variable-name-1" <, "variable-name-2", ...>},

label="string",

lifetime=64-bit-integer,

maxMemSize=64-bit-integer,

memoryFormat="DVR" | "INHERIT" | "STANDARD",

name="table-name",

onDemand=TRUE | FALSE,

promote=TRUE | FALSE,

replace=TRUE | FALSE,

replication=integer,

threadBlockSize=64-bit-integer,

timeStamp="string",

where={"string-1" <, "string-2", ...>}

cdf=TRUE | FALSE,

edfAlpha=double,

maxObsPerModel=integer,

pdf=TRUE | FALSE,

qq=TRUE | FALSE

={{

degree=integer,

details=TRUE | FALSE,

labelStyle

expand=TRUE | FALSE

exponent="string"

includeName=TRUE | FALSE

productSymbol="NONE" | "string"

mDegree=integer,

name="string",

noSeparate=TRUE | FALSE,

standardize

method="MOMENTS" | "MRANGE" | "WMOMENTS"

options="CENTER" | "CENTERSCALE" | "NONE" | "SCALE"

prefix="NONE" | "string"

vars={"variable-name-1" <, "variable-name-2", ...>}

}, {...}},

probObserved=double,

sampleFraction=double,

sampleSize=integer,

effects

={{

interaction="BAR" | "CROSS" | "NONE",

maxInteract=integer,

nest={"string-1" <, "string-2", ...>},

vars={"string-1" <, "string-2", ...>}

}, {...}},

include=integer | {{effect-1} <, {effect-2}, ...>},

informative=TRUE | FALSE,

offset="variable-name",

start=integer | {{effect-1} <, {effect-2}, ...>}

seed=integer,

candidates=integer | "ALL",

choose="AIC" | "AICC" | "DEFAULT" | "NONE" | "SBC",

competitive=TRUE | FALSE,

details="ALL" | "NONE" | "STEPS" | "SUMMARY",

elasticNetOptions

absFConv=double

fConv=double

gConv=double

lambda={double-1 <, double-2, ...>}

mixing={double-1 <, double-2, ...>}

numLambda=integer

rho=double

solver="ADMM" | "BFGS" | "LBFGS" | "NLP"

enscale=TRUE | FALSE,

ensteps=integer,

hierarchy="DEFAULT" | "NONE" | "SINGLE" | "SINGLECLASS",

L2=double,

L2HIGH=double,

L2LOW=double,

maxEffects=integer,

maxSteps=integer,

minEffects=integer,

orderSelect=TRUE | FALSE,

plots=TRUE | FALSE,

select="AIC" | "AICC" | "DEFAULT" | "SBC",

stop="AIC" | "AICC" | "DEFAULT" | "NONE" | "SBC",

stopHorizon=integer

={{

basis="BSPLINE" | "TPF_DEFAULT" | "TPF_NOINT" | "TPF_NOINTANDNOPOWERS" | "TPF_NOPOWERS",

dataBoundary=TRUE | FALSE,

degree=integer,

details=TRUE | FALSE,

knotMax=double,

knotMethod

equal=integer

list={double-1 <, double-2, ...>}

listWithBoundary={double-1 <, double-2, ...>}

multiscale

endScale=integer

startScale=integer

}

rangeFractions={double-1 <, double-2, ...>}

knotMin=double,

name="string",

naturalCubic=TRUE | FALSE,

separate=TRUE | FALSE,

split=TRUE | FALSE,

vars={"variable-name-1" <, "variable-name-2", ...>}

}, {...}},

caslib="string",

compress=TRUE | FALSE,

indexVars={"variable-name-1" <, "variable-name-2", ...>},

label="string",

lifetime=64-bit-integer,

maxMemSize=64-bit-integer,

memoryFormat="DVR" | "INHERIT" | "STANDARD",

name="table-name",

onDemand=TRUE | FALSE,

promote=TRUE | FALSE,

replace=TRUE | FALSE,

replication=integer,

threadBlockSize=64-bit-integer,

timeStamp="string",

where={"string-1" <, "string-2", ...>}

storeCovariance=TRUE | FALSE,

caslib="string",

computedOnDemand=TRUE | FALSE,

computedVars

={{

format="string",

formattedLength=integer,

label="string",

name="variable-name",

nfd=integer,

nfl=integer

}, {...}},

computedVarsProgram="string",

dataSourceOptions={key-1=any-list-or-data-type-1 <, key-2=any-list-or-data-type-2, ...>},

groupBy

={{

format="string",

formattedLength=integer,

label="string",

name="variable-name",

nfd=integer,

nfl=integer

}, {...}},

groupByMode="NOSORT" | "REDISTRIBUTE",

name="table-name",

onDemand=TRUE | FALSE,

orderBy

={{

format="string",

formattedLength=integer,

label="string",

name="variable-name",

nfd=integer,

nfl=integer

}, {...}},

singlePass=TRUE | FALSE,

vars

={{

format="string",

formattedLength=integer,

label="string",

name="variable-name",

nfd=integer,

nfl=integer

}, {...}},

where="where-expression",

whereTable

casLib="string"

name="table-name"

vars

={{

format="string",

formattedLength=integer,

label="string",

name="variable-name",

nfd=integer,

nfl=integer

}, {...}}

where="where-expression"

}

weight="variable-name"

;

indicates a required parameter

Summary: Input and Output Tables

If a row includes a subparameter, you can specify the name, caslib, and so on in the subparameter. Otherwise, you can specify the name, caslib, and so on in the parameter.

Parameters for Reading Input Tables
Parameter	Subparameter	Description
required parameterfuncDef	—	specifies the name of the table that contains the FCMP function definitions.
inest	—	specifies a table to read initial parameter estimates from.
required parametertable	—	specifies the input data table.

Parameters for Creating Output Tables
Parameter	Subparameter	Description
outModelInfo	—	specifies an output table that contains the information about each candidate distribution.
outScoreLibrary	outById	specifies whether and how to create the FCMP code for scoring functions.
outest	required parametertable	specifies whether and how to create a table to write final parameter estimates to.
output	required parametercasOut	specifies the details of the output data table to write scores and quantiles to.
outputTables	names	lists the names of results tables to save as CAS tables on the server.
outstat	—	specifies an output table that contains the values of statistics of fit for each model whose parameter estimation process converges.
store	—	specifies a table that stores the model fit information.

Parameter Descriptions

balanceThreads=TRUE | FALSE

when set to False, turns off the preprocessing that ensures an equitable distribution of the estimation work among threads for each BY group. Setting this parameter to False can cause significantly longer estimation times in some cases, but it can help you obtain numeric results that match the results of SAS Econometrics versions prior to version 8.5.

Default	TRUE

class={{classStatement-1} <, {classStatement-2}, ...>}

specifies the classification variables.

For more information about specifying the class parameter, see the common classStatement parameter (Appendix A: Common Parameters).

Aliases	classVars
Aliases	nominal

classGlobalOpts={classopts}

specifies options to control levelization of the classification variables.

For more information about specifying the classGlobalOpts parameter, see the common classopts parameter (Appendix A: Common Parameters).

classLevelsPrint=TRUE | FALSE

when set to False, suppresses the display of class levels.

Default	TRUE

collectionEffect={{collection-1} <, {collection-2}, ...>}

defines a set of variables that are treated as a single effect that has multiple degrees of freedom.

Alias	collection

The collection value can be one or more of the following:

details=TRUE | FALSE

when set to True, requests a table that shows additional details that are related to this effect.

Default	FALSE

* name="string"

specifies the name of the effect.

* vars={"variable-name-1" <, "variable-name-2", ...>}

specifies a set of variables that are treated as a single effect that has multiple degrees of freedom. The columns in the design matrix that are contributed by a collection effect are the design columns of its constituent variables in the order in which they appear in the definition of the collection effect.

criterion="AD" | "AIC" | "AICC" | "CVM" | "KS" | "LOGLIK" | "SBC"

specifies the criterion to mark the best distribution in the distribution selection table.

Default	LOGLIK

Anderson-Darling statistic

AIC

Akaike's information criterion

AICC

corrected Akaike's information criterion

CVM

Cramer-von Mises statistic

KS

Kolmogorov-Smirnov statistic

LOGLIK

-2 log likelihood

SBC

Schwarz's Bayesian information criterion

Alias	BIC

ctLimits={{ctlim-1} <, {ctlim-2}, ...>}

specifies the global censoring and truncation limits.

The ctlim value can be one or more of the following:

* limit=double

global limit or threshold

* role="LC" | "LT" | "RC" | "RT"

role of the limit

LC

left-censoring

LT

left-truncation

RC

right-censoring

RT

right-truncation

dfMixture={dfmix}

specifies the parameters for computing representative estimates of the cumulative distribution function (CDF) that are used to assess a scale regression model.

Long form	dfMixture={method="FULL" \| "MEAN" \| "QUANTILE" \| "RANDOM"}
Shortcut form	dfMixture="FULL" \| "MEAN" \| "QUANTILE" \| "RANDOM"

The dfmix value can be one or more of the following:

meanType="EXPXBETA" | "XBETA"

specifies how to compute the mean of the scale parameter for the MEAN mixture method.

Default	XBETA

EXPXBETA

computes the scale parameter as mean(exp(log(S_k))), where S_k is the scale parameter of observation k.

XBETA

computes the scale parameter as exp(mean(log(S_k))), where S_k is the scale parameter of observation k.

method="FULL" | "MEAN" | "QUANTILE" | "RANDOM"

specifies which distributions to use in forming the representative mixture distribution.

Default	MEAN

FULL

uses as many distributions as the number of observations.

MEAN

uses one distribution with scale equal to the mean of the scales of all observations.

QUANTILE

uses K-1 distributions each with scale equal to one of the K-1 quantiles of the scales of all observations.

RANDOM

uses K distributions each with scale equal to the scale of one of K randomly chosen observations.

nQuantile=integer

specifies the number of quantiles to use for the QUANTILE mixture method.

Default	2
Minimum value	2

nRandom=integer

specifies the number of points to use for the RANDOM mixture method.

Default	15
Minimum value	1

seed=double

specifies the seed to use for the RANDOM mixture method.

display={displayTables}

specifies a list of results tables to send to the client for display.

For more information about specifying the display parameter, see the common displayTables parameter (Appendix A: Common Parameters).

distributions={"string-1" <, "string-2", ...>}

specifies the list of distribution names to analyze.

empiricalCDF={edfparms}

specifies the parameters for computing the empirical distribution function.

Long form	empiricalCDF={method="AUTO" \| "KAPLANMEIER" \| "MODIFIEDKM" \| "NOTURNBULL" \| "STANDARD" \| "TURNBULL"}
Shortcut form	empiricalCDF="AUTO" \| "KAPLANMEIER" \| "MODIFIEDKM" \| "NOTURNBULL" \| "STANDARD" \| "TURNBULL"

The edfparms value can be one or more of the following:

method="AUTO" | "KAPLANMEIER" | "MODIFIEDKM" | "NOTURNBULL" | "STANDARD" | "TURNBULL"

specifies the method to use for computing the empirical distribution function.

Default	AUTO

AUTO

automatically selects the best method based on data.

KAPLANMEIER

uses the Kaplan-Meier estimation method.

MODIFIEDKM

uses a modified Kaplan-Meier estimation method that ignores contributions from observations with very small risk sets.

NOTURNBULL

automatically selects the best non-Turnbull method. This essentially disables Turnbull's method, and interval-censored observations are approximated by the midpoint of the interval.

STANDARD

uses a standard empirical CDF that ignores censoring and truncation.

TURNBULL

uses Turnbull's method. This is used only when both left-censoring and right-censoring are specified.

riskSetLowerBound=double

specifies the lower bound on risk set size. This applies only to the modified Kaplan-Meier method.

Alias	rslb

riskSetLowerBoundAlpha=double

specifies the value to use for Alpha to compute the lower bound on the risk set size as C*(N**Alpha). This applies only to the modified Kaplan-Meier method.

Alias	alpha
Default	0.5
Range	(0, 1)

riskSetLowerBoundC=double

specifies the value to use for C to compute the lower bound on the risk set size as C*(N**Alpha). This applies only to the modified Kaplan-Meier method.

Alias	c
Default	1
Minimum value (exclusive)	0

turnbullEnsureMLE=TRUE | FALSE

specifies that the final empirical distribution function estimates be maximum likelihood (ML) estimates, because the expectation-maximization algorithm might not always converge at ML estimates. This applies only to the Turnbull method.

Alias	ensureMLE
Default	FALSE

turnbullMaxError=double

specifies the maximum relative error to be allowed between estimates of two consecutive iterations. This applies only to the Turnbull method.

Alias	eps
Default	1E-08

turnbullMaxIter=integer

specifies the maximum number of iterations to attempt to find the empirical estimates. This applies only to the Turnbull method.

Alias	maxiter
Default	500

turnbullZeroProb=double

specifies the threshold below which an empirical estimate of the probability is considered 0. This applies only to the Turnbull method when you request that final estimates be maximum likelihood estimates.

Alias	zeroprob
Default	1E-08

estimation={nlopts}

specifies parameters that control various aspects of the parameter estimation process.

Alias	nloptions

Long form	estimation={technique="CONGRA" \| "DBLDOG" \| "DUQUAN" \| "NEWRAP" \| "NMSIMP" \| "NONE" \| "NRRIDG" \| "QUANEW" \| "TRUREG"}
Shortcut form	estimation="CONGRA" \| "DBLDOG" \| "DUQUAN" \| "NEWRAP" \| "NMSIMP" \| "NONE" \| "NRRIDG" \| "QUANEW" \| "TRUREG"

The nlopts value can be one or more of the following:

absConv=double

specifies an absolute function convergence criterion.

Alias	absTol
Minimum value (exclusive)	0

absFconv=double

specifies an absolute function difference convergence criterion.

Alias	absFtol
Minimum value (exclusive)	0

absGconv=double

specifies an absolute gradient convergence criterion.

Alias	absGtol
Minimum value (exclusive)	0

absXconv=double

specifies an absolute parameter convergence criterion.

Alias	absXtol
Minimum value (exclusive)	0

fConv=double

specifies a relative function convergence criterion.

Alias	fTol
Minimum value (exclusive)	0

fConv2=double

specifies a second function convergence criterion.

Alias	fTol2
Minimum value (exclusive)	0

fsize=double

specifies the FSIZE parameter of the relative function and relative gradient termination criteria.

Minimum value (exclusive)	0

gConv=double

specifies a relative gradient convergence criterion.

Alias	gTol
Minimum value (exclusive)	0

gConv2=double

specifies another relative gradient convergence criterion.

Alias	gTol2
Minimum value (exclusive)	0

maxFunc=double

specifies the maximum number of objective function evaluations in the optimization process.

Minimum value	0

maxIter=double

specifies the maximum number of iterations in the optimization process. The default varies by the technique.

Minimum value	0

maxTime=double

specifies an upper limit (in seconds) on the CPU time for the optimization process.

Minimum value (exclusive)	0

minIter=integer

specifies the minimum number of iterations in the optimization process.

Minimum value	0

restart=double

specifies the number of iterations after which the QUANEW or CONGRA technique is restarted with a steepest search direction.

Minimum value	1

singular=double

specifies the singularity criterion that is used for the inversion of the Hessian matrix.

Range	(0, 1)

technique="CONGRA" | "DBLDOG" | "DUQUAN" | "NEWRAP" | "NMSIMP" | "NONE" | "NRRIDG" | "QUANEW" | "TRUREG"

specifies the nonlinear optimization technique to use.

Default	TRUREG

CONGRA

conjugate-gradient

DBLDOG

double-dogleg

DUQUAN

dual quasi-Newton

NEWRAP

Newton-Raphson (line search and ridging)

NMSIMP

Nelder-Mead simplex

NONE

no optimization

NRRIDG

Newton-Raphson with ridging

QUANEW

quasi-Newton

TRUREG

trust region

techniqueSelect="CONGRA" | "DBLDOG" | "DUQUAN" | "NEWRAP" | "NMSIMP" | "NONE" | "NRRIDG" | "QUANEW" | "TRUREG"

specifies the nonlinear optimization technique to use for intermediate scale model selection steps.

CONGRA

conjugate-gradient

DBLDOG

double-dogleg

DUQUAN

dual quasi-Newton

NEWRAP

Newton-Raphson (line search and ridging)

NMSIMP

Nelder-Mead simplex

NONE

no optimization

NRRIDG

Newton-Raphson with ridging

QUANEW

quasi-Newton

TRUREG

trust region

varianceDivisor="DF" | "N"

specifies the denominator to use for computing the covariance estimates.

Alias	vardef
Default	DF

DF

uses degrees of freedom as the denominator.

N

uses the number of nonmissing observations as the denominator.

xConv=double

specifies the relative parameter convergence criterion.

Alias	xTol
Minimum value (exclusive)	0

xsize=double

specifies the XSIZE parameter r of the relative parameter termination criterion.

Minimum value (exclusive)	0

freq="variable-name"

specifies the observation frequency variable.

* funcDef={castable}

specifies the name of the table that contains the FCMP function definitions.

Long form	funcDef={name="table-name"}
Shortcut form	funcDef="table-name"

The castable value can be one or more of the following:

caslib="string"

specifies the caslib for the input table that you want to use with the action. By default, the active caslib is used. Specify a value only if you need to access a table from a different caslib.

dataSourceOptions={key-1=any-list-or-data-type-1 <, key-2=any-list-or-data-type-2, ...>}

specifies data source options.

Aliases	options
Aliases	dataSource

* name="table-name"

specifies the name of the input table.

whereTable={groupbytable}

specifies an input table that contains rows to use as a WHERE filter. If the vars parameter is not specified, then all the variable names that are common to the input table and the filtering table are used to find matching rows. If the where parameter for the input table and this parameter are specified, then this filtering table is applied first.

The groupbytable value can be one or more of the following:

casLib="string"

specifies the caslib for the filter table. By default, the active caslib is used.

dataSourceOptions={adls_noreq-parameters | bigquery-parameters | cas_noreq-parameters | clouddex-parameters | db2-parameters | dnfs-parameters | esp-parameters | fedsvr-parameters | gcs_noreq-parameters | hadoop-parameters | hana-parameters | impala-parameters | jdbc-parameters | mongodb-parameters | mysql-parameters | odbc-parameters | oracle-parameters | path-parameters | postgres-parameters | redshift-parameters | s3-parameters | sapiq-parameters | sforce-parameters | snowflake-parameters | spark-parameters | spde-parameters | sqlserver-parameters | ss_noreq-parameters | teradata-parameters | vertica-parameters | yellowbrick-parameters}

specifies data source options.

Aliases	options
Aliases	dataSource

For more information about specifying the dataSourceOptions parameter, see the common dataSourceOptions parameter (Appendix A: Common Parameters).

importOptions={fileType="ANY" | "AUDIO" | "AUTO" | "BASESAS" | "CSV" | "DOCUMENT" | "DTA" | "ESP" | "EXCEL" | "FMT" | "HDAT" | "IMAGE" | "JMP" | "LASR" | "PARQUET" | "SPSS" | "VIDEO" | "XLS", fileType-specific-parameters}

specifies the settings for reading a table from a data source.

Alias	import

For more information about specifying the importOptions parameter, see the common importOptions parameter (Appendix A: Common Parameters).

* name="table-name"

specifies the name of the filter table.

vars={{casinvardesc-1} <, {casinvardesc-2}, ...>}

specifies the variable names to use from the filter table.

The casinvardesc value can be one or more of the following:

format="string"

specifies the format to apply to the variable.

formattedLength=integer

specifies the length of format field plus the length of the format precision.

label="string"

specifies the descriptive label for the variable.

* name="variable-name"

specifies the name for the variable.

nfd=integer

specifies the length of the format precision.

nfl=integer

specifies the length of the format field.

where="where-expression"

specifies an expression for subsetting the data from the filter table.

inest={castable}

specifies a table to read initial parameter estimates from.

Long form	inest={name="table-name"}
Shortcut form	inest="table-name"

The castable value can be one or more of the following:

caslib="string"

specifies the caslib for the input table that you want to use with the action. By default, the active caslib is used. Specify a value only if you need to access a table from a different caslib.

dataSourceOptions={key-1=any-list-or-data-type-1 <, key-2=any-list-or-data-type-2, ...>}

specifies data source options.

Aliases	options
Aliases	dataSource

groupBy={{casinvardesc-1} <, {casinvardesc-2}, ...>}

specifies the names of the variables to use for grouping results.

The casinvardesc value can be one or more of the following:

format="string"

specifies the format to apply to the variable.

formattedLength=integer

specifies the length of format field plus the length of the format precision.

label="string"

specifies the descriptive label for the variable.

* name="variable-name"

specifies the name for the variable.

nfd=integer

specifies the length of the format precision.

nfl=integer

specifies the length of the format field.

groupByMode="NOSORT" | "REDISTRIBUTE"

specifies how to create groups.

Default	NOSORT

NOSORT

group the data without sorting on each machine, and then groups the data again on the controller.

REDISTRIBUTE

transfers rows between nodes to guarantee ordering within groups. This method is slower.

importOptions={fileType="ANY" | "AUDIO" | "AUTO" | "BASESAS" | "CSV" | "DOCUMENT" | "DTA" | "ESP" | "EXCEL" | "FMT" | "HDAT" | "IMAGE" | "JMP" | "LASR" | "PARQUET" | "SPSS" | "VIDEO" | "XLS", fileType-specific-parameters}

specifies the settings for reading a table from a data source.

Alias	import

For more information about specifying the importOptions parameter, see the common importOptions parameter (Appendix A: Common Parameters).

* name="table-name"

specifies the name of the input table.

where="where-expression"

specifies an expression for subsetting the input data.

whereTable={groupbytable}

The groupbytable value can be one or more of the following:

casLib="string"

specifies the caslib for the filter table. By default, the active caslib is used.

dataSourceOptions={adls_noreq-parameters | bigquery-parameters | cas_noreq-parameters | clouddex-parameters | db2-parameters | dnfs-parameters | esp-parameters | fedsvr-parameters | gcs_noreq-parameters | hadoop-parameters | hana-parameters | impala-parameters | jdbc-parameters | mongodb-parameters | mysql-parameters | odbc-parameters | oracle-parameters | path-parameters | postgres-parameters | redshift-parameters | s3-parameters | sapiq-parameters | sforce-parameters | snowflake-parameters | spark-parameters | spde-parameters | sqlserver-parameters | ss_noreq-parameters | teradata-parameters | vertica-parameters | yellowbrick-parameters}

specifies data source options.

Aliases	options
Aliases	dataSource

For more information about specifying the dataSourceOptions parameter, see the common dataSourceOptions parameter (Appendix A: Common Parameters).

importOptions={fileType="ANY" | "AUDIO" | "AUTO" | "BASESAS" | "CSV" | "DOCUMENT" | "DTA" | "ESP" | "EXCEL" | "FMT" | "HDAT" | "IMAGE" | "JMP" | "LASR" | "PARQUET" | "SPSS" | "VIDEO" | "XLS", fileType-specific-parameters}

specifies the settings for reading a table from a data source.

Alias	import

For more information about specifying the importOptions parameter, see the common importOptions parameter (Appendix A: Common Parameters).

* name="table-name"

specifies the name of the filter table.

vars={{casinvardesc-1} <, {casinvardesc-2}, ...>}

specifies the variable names to use from the filter table.

The casinvardesc value can be one or more of the following:

format="string"

specifies the format to apply to the variable.

formattedLength=integer

specifies the length of format field plus the length of the format precision.

label="string"

specifies the descriptive label for the variable.

* name="variable-name"

specifies the name for the variable.

nfd=integer

specifies the length of the format precision.

nfl=integer

specifies the length of the format field.

where="where-expression"

specifies an expression for subsetting the data from the filter table.

initialDistributionParameters={{initdist-1} <, {initdist-2}, ...>}

specifies initial values of distribution parameters.

Alias	initVals

The initdist value can be one or more of the following:

* distribution="string"

specifies the distribution name.

Alias	dist

* parameters={{distparm-1} <, {distparm-2}, ...>}

specifies the list of name-value pairs for the parameters that you want to initialize.

Aliases	parm
Aliases	parms

The distparm value can be one or more of the following:

* name="string"

specifies the name of the distribution parameter that you want to initialize.

* value=double

specifies the initial value for the specified distribution parameter.

lossVariables={{lossrole-1} <, {lossrole-2}, ...>}

specifies variables related to the loss response, such as target, censoring, and truncation.

The lossrole value can be one or more of the following:

* name="variable-name"

specifies the name of the variable (must be present in the input table).

* role="LC" | "LT" | "RC" | "RT" | "TARGET"

specifies the role of the variable.

LC

specifies the left-censoring variable name.

LT

specifies the left-truncation variable name.

RC

specifies the right-censoring variable name.

RT

specifies the right-truncation variable name.

TARGET

specifies the loss (dependent or target) variable name.

multimemberEffect={{multimember-1} <, {multimember-2}, ...>}

uses one or more classification variables specified in the vars parameter in such a way that each observation can be associated with one or more levels of the union of the levels of the classification variables.

For more information about specifying the multimemberEffect parameter, see the common multimember parameter (Appendix A: Common Parameters).

Aliases	multimember
Aliases	mmEffect

nClassLevelsPrint=integer

limits the display of class levels. The value 0 suppresses all levels.

Minimum value	0

noConstFitStats=TRUE | FALSE

when set to True, excludes any constant distribution parameters from the calculations of likelihood-based fit statistics.

Alias	noConstSOF
Default	FALSE

objective={objSpec}

specifies a custom objective function. If you do not specify this parameter, the default objective function, which is the negative of the log likelihood, is used.

Alias	obj

The objSpec value can be one or more of the following:

* SASProgram="string"

specifies programming statements that compute the objective function. The computed value is minimized.

Aliases	SASCode
Aliases	code

* symbol="string"

specifies the objective function symbol.

Alias	sym

outDetailLevel=integer

specifies the level of detail to be printed to the output tables.

Default	1
Minimum value	0

outest={outest}

specifies whether and how to create a table to write final parameter estimates to.

The outest value can be one or more of the following:

covout=TRUE | FALSE

when set to True, writes covariance estimates to the OUTEST table.

Default	FALSE

selectOut=TRUE | FALSE

when set to True, writes only the regression parameters that correspond to the selected effects to the OUTEST table.

Default	FALSE

* table={casouttable}

specifies a table to write final parameter estimates to.

For more information about specifying the table parameter, see the common casouttable parameter (Appendix A: Common Parameters).

zeroEstForNotInModel=TRUE | FALSE

when set to True, writes 0 to the OUTEST table as the estimate of the regression parameter not in the model (because it is either collinear or not selected). When set to False, writes a missing value.

Alias	zeroEst
Default	FALSE

outModelInfo={casouttable}

specifies an output table that contains the information about each candidate distribution.

For more information about specifying the outModelInfo parameter, see the common casouttable parameter (Appendix A: Common Parameters).

output={sevOutputStatement}

specifies the details of the output data table to write scores and quantiles to.

The sevOutputStatement value can be one or more of the following:

* casOut={casouttable}

specifies the settings for an output table.

For more information about specifying the casOut parameter, see the common casouttable parameter (Appendix A: Common Parameters).

copyVars="ALL" | "ALL_MODEL" | "ALL_NUMERIC" | {"variable-name-1" <, "variable-name-2", ...>}

specifies a list of one or more variables to be copied from the input table to the output table. You can alternatively specify the value ALL, ALL_MODEL, or ALL_NUMERIC, which respectively copies all variables, all variables used in the modeling, or all numeric variables from the input table to the output table.

quantiles={outquant}

specifies the quantiles to write to the output data table for each distribution.

The outquant value can be one or more of the following:

names={"string-1" <, "string-2", ...>}

specifies the names of the quantile variables. If this list of names is shorter than the list of CDF values, then quantile variables for the remaining CDF values get default names.

nDecimal=integer

specifies the numeric precision (number of digits after the decimal point) to use for creating default names of the quantile variables.

Minimum value	0

* points={{points-1} <, {points-2}, ...>}

specifies the CDF values to evaluate the quantile function at.

Alias	cdfValues

* p=double

Range	(0, 1)

scoreFunctions={{sfunc-1} <, {sfunc-2}, ...>}

specifies the scoring functions to evaluate and write to the output data table.

Alias	scores

The sfunc value can be one or more of the following:

arguments={double-1 <, double-2, ...>}

specifies a value for the first argument of the scoring function. If this value is not specified, then the response variable's value in the input data table is used.

Alias	arg

* name="string"

specifies the name of the scoring function. For each converged model based on distribution D, function D_<name> is evaluated and written as a variable in the output data table.

variable="string"

specifies the name of the column for the scoring function (default is the name of the scoring function itself).

Alias	var

outputTables={outputTables}

lists the names of results tables to save as CAS tables on the server.

For more information about specifying the outputTables parameter, see the common outputTables parameter (Appendix A: Common Parameters).

Alias	displayOut

outScoreLibrary={outscorelib}

specifies whether and how to create the FCMP code for scoring functions.

Alias	outscorelib

The outscorelib value can be one or more of the following:

commonPackage=TRUE | FALSE

when set to True, creates only one common package to contain all the scoring functions.

Alias	onePackage
Default	FALSE

outById={casouttable}

specifies an output table that contains the unique identifier for each BY group. This is a required parameter when you request BY-group processing.

For more information about specifying the outById parameter, see the common casouttable parameter (Appendix A: Common Parameters).

outstat={casouttable}

specifies an output table that contains the values of statistics of fit for each model whose parameter estimation process converges.

For more information about specifying the outstat parameter, see the common casouttable parameter (Appendix A: Common Parameters).

plotTable={plotTable}

specifies the information to write to the PlotData results table. You can use this table to prepare plots.

Alias	plot

The plotTable value can be one or more of the following:

cdf=TRUE | FALSE

when set to True, includes the cumulative distribution function (CDF) information in the PlotData table.

Default	FALSE

edfAlpha=double

specifies the confidence level to use for computing the confidence intervals for the empirical distribution function (EDF) estimates.

Default	0.05
Range	(0, 1)

maxObsPerModel=integer

specifies the maximum number of observations per model in the PlotData table. If the size of the sample that is used for computing the fit statistics is larger than this number, the PlotData table is not generated. Because this limit applies to each model, the total number of observations in the PlotData table for each BY group is equal to the product of this number and the number of models that do not fail to converge. If you specify a large number, a large amount of data is communicated to the client, which can cost time and money without much gain in the discernibility of the plots.

Alias	maxObs
Default	5000
Minimum value	5

pdf=TRUE | FALSE

when set to True, includes the probability density function (PDF) information in the PlotData table.

Default	FALSE

qq=TRUE | FALSE

when set to True, includes the quantile information in the PlotData table.

Default	FALSE

polynomialEffect={{polynomial-1} <, {polynomial-2}, ...>}

specifies a polynomial effect. All specified variables must be numeric. A design matrix column is generated for each term of the specified polynomial. By default, each of these terms is treated as a separate effect for the purpose of model building.

For more information about specifying the polynomialEffect parameter, see the common polynomial parameter (Appendix A: Common Parameters).

Aliases	poly
Aliases	polynomial

probObserved=double

specifies the probability of observability to use with left-truncation specification.

Range	(0, 1)

sampleFraction=double

specifies the fraction of observations to use to form a random sample, which is used to compute initial parameter values and EDF-based fit statistics.

Alias	initFraction
Range	0–1

sampleSize=integer

specifies the approximate number of observations to use per node to form a random sample, which is used to compute initial parameter values and EDF-based fit statistics.

Alias	initSize
Default	10000
Minimum value	50

scaleModel={modelStatement}

specifies the effects to consider for the scale regression model.

The modelStatement value can be one or more of the following:

effects={{effect-1} <, {effect-2}, ...>}

specifies a list of effects that define the model. Each term in this list is made up of variables specified in the vars parameter and their interaction (which can be NONE, CROSS, or BAR). When the interaction is BAR, it can be limited by the maxInteract parameter.

The effect value can be one or more of the following:

interaction="BAR" | "CROSS" | "NONE"

specifies the type of interaction for the variables.

Alias	interact
Default	NONE

maxInteract=integer

eliminates interaction effects whose order is higher than the specified integer value when used in conjunction with the BAR interaction.

nest={"string-1" <, "string-2", ...>}

specifies the variables to be nested within the term that is defined by the vars parameter. For terms with a BAR or CROSS interaction, the nest corresponds to the last variable in the vars parameter. For terms with no interaction, the nest is distributed across all variables that are listed in the vars parameter.

* vars={"string-1" <, "string-2", ...>}

specifies the variables to use in defining a term of the effect. You must specify at least one variable.

include=integer | {{effect-1} <, {effect-2}, ...>}

specifies effects to include at the start of the selection process for the specified selection method. Included effects are never dropped during the selection process. If you specify n, where n is a positive integer, then the included effects consist of the first n effects of the model specification.

The effect value is specified as follows:

interaction="BAR" | "CROSS" | "NONE"

specifies the type of interaction for the variables.

Alias	interact
Default	NONE

maxInteract=integer

eliminates interaction effects whose order is higher than the specified integer value when used in conjunction with the BAR interaction.

nest={"string-1" <, "string-2", ...>}

* vars={"string-1" <, "string-2", ...>}

specifies the variables to use in defining a term of the effect. You must specify at least one variable.

informative=TRUE | FALSE

when set to True, models missing values by using extra model effects. These effects consist of dummy variables that take the value 1 when the value of a continuous model variable involved in the effect is missing, and take the value 0 otherwise. The missing value in the original model effect is replaced by the average value of the effect for the nonmissing values. For classification variables, missing values are treated as valid levels.

Default	FALSE

offset="variable-name"

specifies a numeric offset variable. This variable cannot be a classification variable, a response variable, or one of the explanatory variables.

start=integer | {{effect-1} <, {effect-2}, ...>}

specifies effects to use to begin the selection process in the FORWARD, FORWARDSWAP, and STEPWISE selection methods. If you specify n, where n is a positive integer, then the starting model consists of the first n effects of the model specification.

The effect value is specified as follows:

interaction="BAR" | "CROSS" | "NONE"

specifies the type of interaction for the variables.

Alias	interact
Default	NONE

maxInteract=integer

eliminates interaction effects whose order is higher than the specified integer value when used in conjunction with the BAR interaction.

nest={"string-1" <, "string-2", ...>}

* vars={"string-1" <, "string-2", ...>}

specifies the variables to use in defining a term of the effect. You must specify at least one variable.

seed=integer

specifies the seed for random selection of initialization sample.

Minimum value	0

selection={selectionStatement}

specifies scale model selection parameters.

Long form	selection={method="BACKWARD" \| "ELASTICNET" \| "FORWARD" \| "FORWARDSWAP" \| "NONE" \| "STEPWISE"}
Shortcut form	selection="BACKWARD" \| "ELASTICNET" \| "FORWARD" \| "FORWARDSWAP" \| "NONE" \| "STEPWISE"

The selectionStatement value can be one or more of the following:

candidates=integer | "ALL"

specifies the maximum number of candidates to display at each step of the selection process, when the detail level ALL is specified.

choose="AIC" | "AICC" | "DEFAULT" | "NONE" | "SBC"

specifies the criterion for choosing the model. The specified criterion is evaluated at each step of the selection process, and the model that yields the best value of the criterion is chosen.

competitive=TRUE | FALSE

when set to True, evaluates (during stepwise selection) the selection criterion for all models in which an effect currently in the model is dropped or an effect not yet in the model is added. The effect whose removal from or addition to the model yields the maximum improvement to the selection criterion is dropped or added.

Default	FALSE

details="ALL" | "NONE" | "STEPS" | "SUMMARY"

specifies the level of detail to produce about the selection process.

Default	SUMMARY

elasticNetOptions={enOptions}

specifies options to use in performing elastic net selection methods.

The enOptions value can be one or more of the following:

absFConv=double

specifies the absolute function difference convergence criterion.

Alias	abstol
Default	1E-08
Minimum value	0

fConv=double

specifies the relative function difference convergence criterion.

Alias	fTol
Minimum value	0

gConv=double

specifies the relative gradient convergence criterion.

Alias	gTol
Minimum value	0

lambda={double-1 <, double-2, ...>}

specifies the regularization parameters in the elastic net selection method.

mixing={double-1 <, double-2, ...>}

specifies the elastic net mixing parameter.

numLambda=integer

specifies the number of regularization parameters in the elastic net selection method.

Alias	nLambda
Default	0
Minimum value	0

rho=double

specifies the scaling factor to use in computing minimum regularization parameter.

Range	(0, 1)

solver="ADMM" | "BFGS" | "LBFGS" | "NLP"

specifies a solver for elastic net selection.

enscale=TRUE | FALSE

when set to True, applies scaling to beta in the elastic net selection method.

Default	FALSE

ensteps=integer

specifies the number of iterations to use in the elastic net selection method.

Default	50

hierarchy="DEFAULT" | "NONE" | "SINGLE" | "SINGLECLASS"

specifies whether and how to apply the model hierarchy requirement. Model hierarchy refers to the requirement that, for any term to be in the model, all model effects that are contained in the term must be present in the model.

Default	DEFAULT

L2=double

specifies the L2 parameter in the elastic net selection method.

Default	0

L2HIGH=double

specifies the upper bound to use in searching for the L2 parameter in the elastic net selection method.

Alias	maxL2
Default	1

L2LOW=double

specifies the lower bound to use in searching for the L2 parameter in the elastic net selection method.

Alias	minL2
Default	0

maxEffects=integer

specifies the maximum number of effects in any model to consider during the selection process. This parameter is ignored for backward selection.

maxSteps=integer

specifies the maximum number of selection steps to perform.

method="BACKWARD" | "ELASTICNET" | "FORWARD" | "FORWARDSWAP" | "NONE" | "STEPWISE"

specifies the model selection method.

Default	STEPWISE

minEffects=integer

specifies the minimum number of effects in any model to consider during backward selection.

orderSelect=TRUE | FALSE

when set to True, shows effects in parameter estimates tables in the order in which they were added to the model.

Default	FALSE

plots=TRUE | FALSE

when set to True, produces coefficientProgression and selectionSummaryForPlots tables that you can use to produce selection diagnostic plots.

Default	FALSE

select="AIC" | "AICC" | "DEFAULT" | "SBC"

specifies the criterion to use in determining the order in which effects enter or leave at each step of the selection method. This parameter does not apply to LAR or LASSO selection.

stop="AIC" | "AICC" | "DEFAULT" | "NONE" | "SBC"

specifies a criterion that to use for stopping the selection process. If you do not specify a stop criterion, then the criterion that is used to select the model is also used as the stop criterion.

stopHorizon=integer

specifies the number of consecutive steps at which the stop criterion must worsen in order for a local extremum to be detected.

Default	3

selectionDistParmInit="FIRST" | "FULL" | "PER"

specifies the method of initializing the distribution parameters for the regression effect selection process.

Alias	slctDistInit
Default	FIRST

FIRST

initializes the distribution parameters once, based on the first candidate model.

FULL

initializes the distribution parameters once, based on the full model.

PER

reinitializes the distribution parameters for every candidate model.

splineEffect={{spline-1} <, {spline-2}, ...>}

expands variables into spline bases whose form depends on the specified parameters.

For more information about specifying the splineEffect parameter, see the common spline parameter (Appendix A: Common Parameters).

Alias	spline

store={casouttable}

specifies a table that stores the model fit information.

For more information about specifying the store parameter, see the common casouttable parameter (Appendix A: Common Parameters).

storeCovariance=TRUE | FALSE

when set to True, writes the parameter covariance estimates to the store that you specify in the store parameter.

Alias	storeCov
Default	TRUE

* table={castable}

specifies the input data table.

For more information about specifying the table parameter, see the common castable parameter (Appendix A: Common Parameters).

weight="variable-name"

specifies the observation weight variable.

severity Action

Estimates parameters of the specified severity distribution models.

Lua Syntax

Expand all Collapse all

results, info = s:severity_severity{

balanceThreads=true | false,

class

={{

countMissing=true | false,

descending=true | false,

ignoreMissing=true | false,

levelizeRaw=true | false,

maxLev=integer,

order="FORMATTED" | "FREQ" | "FREQFORMATTED" | "FREQINTERNAL" | "INTERNAL",

ref="FIRST" | "LAST" | double | "string",

split=true | false,

vars={"variable-name-1" <, "variable-name-2", ...>}

}, {...}},

classGlobalOpts

countMissing=true | false,

descending=true | false,

ignoreMissing=true | false,

levelizeRaw=true | false,

maxLev=integer,

order="FORMATTED" | "FREQ" | "FREQFORMATTED" | "FREQINTERNAL" | "INTERNAL",

ref="FIRST" | "LAST" | double | "string",

split=true | false

classLevelsPrint=true | false,

collectionEffect

={{

details=true | false,

name="string",

vars={"variable-name-1" <, "variable-name-2", ...>}

}, {...}},

criterion="AD" | "AIC" | "AICC" | "CVM" | "KS" | "LOGLIK" | "SBC",

ctLimits

={{

limit=double,

role="LC" | "LT" | "RC" | "RT"

}, {...}},

dfMixture

meanType="EXPXBETA" | "XBETA",

method="FULL" | "MEAN" | "QUANTILE" | "RANDOM",

nQuantile=integer,

nRandom=integer,

seed=double

display

caseSensitive=true | false,

exclude=true | false,

excludeAll=true | false,

keyIsPath=true | false,

names={"string-1" <, "string-2", ...>},

pathType="LABEL" | "NAME",

traceNames=true | false

distributions={"string-1" <, "string-2", ...>},

empiricalCDF

riskSetLowerBound=double,

riskSetLowerBoundAlpha=double,

riskSetLowerBoundC=double,

turnbullEnsureMLE=true | false,

turnbullMaxError=double,

turnbullMaxIter=integer,

turnbullZeroProb=double

estimation

absConv=double,

absFconv=double,

absGconv=double,

absXconv=double,

fConv=double,

fConv2=double,

fsize=double,

gConv=double,

gConv2=double,

maxFunc=double,

maxIter=double,

maxTime=double,

minIter=integer,

restart=double,

singular=double,

varianceDivisor="DF" | "N",

xConv=double,

xsize=double

freq="variable-name",

funcDef

caslib="string",

dataSourceOptions={key-1=any-list-or-data-type-1 <, key-2=any-list-or-data-type-2, ...>},

name="table-name",

casLib="string"

name="table-name"

={{

format="string",

formattedLength=integer,

label="string",

name="variable-name",

nfd=integer,

nfl=integer

}, {...}}

where="where-expression"

}

caslib="string",

dataSourceOptions={key-1=any-list-or-data-type-1 <, key-2=any-list-or-data-type-2, ...>},

groupBy

={{

format="string",

formattedLength=integer,

label="string",

name="variable-name",

nfd=integer,

nfl=integer

}, {...}},

groupByMode="NOSORT" | "REDISTRIBUTE",

name="table-name",

where="where-expression",

casLib="string"

name="table-name"

={{

format="string",

formattedLength=integer,

label="string",

name="variable-name",

nfd=integer,

nfl=integer

}, {...}}

where="where-expression"

}

={{

distribution="string",

parameters

={{

name="string",

value=double

}, {...}}

}, {...}},

={{

name="variable-name",

role="LC" | "LT" | "RC" | "RT" | "TARGET"

}, {...}},

={{

details=true | false,

name="string",

noEffect=true | false,

stdize=true | false,

vars={"variable-name-1" <, "variable-name-2", ...>},

weight={"variable-name-1" <, "variable-name-2", ...>}

}, {...}},

nClassLevelsPrint=integer,

noConstFitStats=true | false,

SASProgram="string",

symbol="string"

outDetailLevel=integer,

covout=true | false,

selectOut=true | false,

caslib="string"

compress=true | false

indexVars={"variable-name-1" <, "variable-name-2", ...>}

label="string"

lifetime=64-bit-integer

maxMemSize=64-bit-integer

memoryFormat="DVR" | "INHERIT" | "STANDARD"

name="table-name"

onDemand=true | false

promote=true | false

replace=true | false

replication=integer

threadBlockSize=64-bit-integer

timeStamp="string"

where={"string-1" <, "string-2", ...>}

zeroEstForNotInModel=true | false

caslib="string",

compress=true | false,

indexVars={"variable-name-1" <, "variable-name-2", ...>},

label="string",

lifetime=64-bit-integer,

maxMemSize=64-bit-integer,

memoryFormat="DVR" | "INHERIT" | "STANDARD",

name="table-name",

onDemand=true | false,

promote=true | false,

replace=true | false,

replication=integer,

threadBlockSize=64-bit-integer,

timeStamp="string",

where={"string-1" <, "string-2", ...>}

casOut

caslib="string"

compress=true | false

indexVars={"variable-name-1" <, "variable-name-2", ...>}

label="string"

lifetime=64-bit-integer

maxMemSize=64-bit-integer

memoryFormat="DVR" | "INHERIT" | "STANDARD"

name="table-name"

onDemand=true | false

promote=true | false

replace=true | false

replication=integer

threadBlockSize=64-bit-integer

timeStamp="string"

where={"string-1" <, "string-2", ...>}

copyVars="ALL" | "ALL_MODEL" | "ALL_NUMERIC" | {"variable-name-1" <, "variable-name-2", ...>},

quantiles

names={"string-1" <, "string-2", ...>}

nDecimal=integer

points

={{

p=double

}, {...}}

scoreFunctions

={{

arguments={double-1 <, double-2, ...>},

name="string",

variable="string"

}, {...}}

groupByVarsRaw=true | false,

includeAll=true | false,

names={"string-1" <, "string-2", ...>} | {key-1={casouttable-1} <, key-2={casouttable-2}, ...>},

repeated=true | false,

replace=true | false

commonPackage=true | false,

outById

caslib="string"

compress=true | false

indexVars={"variable-name-1" <, "variable-name-2", ...>}

label="string"

lifetime=64-bit-integer

maxMemSize=64-bit-integer

memoryFormat="DVR" | "INHERIT" | "STANDARD"

name="table-name"

onDemand=true | false

promote=true | false

replace=true | false

replication=integer

threadBlockSize=64-bit-integer

timeStamp="string"

where={"string-1" <, "string-2", ...>}

}

caslib="string",

compress=true | false,

indexVars={"variable-name-1" <, "variable-name-2", ...>},

label="string",

lifetime=64-bit-integer,

maxMemSize=64-bit-integer,

memoryFormat="DVR" | "INHERIT" | "STANDARD",

name="table-name",

onDemand=true | false,

promote=true | false,

replace=true | false,

replication=integer,

threadBlockSize=64-bit-integer,

timeStamp="string",

where={"string-1" <, "string-2", ...>}

cdf=true | false,

edfAlpha=double,

maxObsPerModel=integer,

pdf=true | false,

qq=true | false

={{

degree=integer,

details=true | false,

labelStyle

expand=true | false

exponent="string"

includeName=true | false

productSymbol="NONE" | "string"

mDegree=integer,

name="string",

noSeparate=true | false,

standardize

method="MOMENTS" | "MRANGE" | "WMOMENTS"

options="CENTER" | "CENTERSCALE" | "NONE" | "SCALE"

prefix="NONE" | "string"

vars={"variable-name-1" <, "variable-name-2", ...>}

}, {...}},

probObserved=double,

sampleFraction=double,

sampleSize=integer,

effects

={{

interaction="BAR" | "CROSS" | "NONE",

maxInteract=integer,

nest={"string-1" <, "string-2", ...>},

vars={"string-1" <, "string-2", ...>}

}, {...}},

include=integer | {{effect-1} <, {effect-2}, ...>},

informative=true | false,

offset="variable-name",

start=integer | {{effect-1} <, {effect-2}, ...>}

seed=integer,

candidates=integer | "ALL",

choose="AIC" | "AICC" | "DEFAULT" | "NONE" | "SBC",

competitive=true | false,

details="ALL" | "NONE" | "STEPS" | "SUMMARY",

elasticNetOptions

absFConv=double

fConv=double

gConv=double

lambda={double-1 <, double-2, ...>}

mixing={double-1 <, double-2, ...>}

numLambda=integer

rho=double

solver="ADMM" | "BFGS" | "LBFGS" | "NLP"

enscale=true | false,

ensteps=integer,

hierarchy="DEFAULT" | "NONE" | "SINGLE" | "SINGLECLASS",

L2=double,

L2HIGH=double,

L2LOW=double,

maxEffects=integer,

maxSteps=integer,

minEffects=integer,

orderSelect=true | false,

plots=true | false,

select="AIC" | "AICC" | "DEFAULT" | "SBC",

stop="AIC" | "AICC" | "DEFAULT" | "NONE" | "SBC",

stopHorizon=integer

={{

basis="BSPLINE" | "TPF_DEFAULT" | "TPF_NOINT" | "TPF_NOINTANDNOPOWERS" | "TPF_NOPOWERS",

dataBoundary=true | false,

degree=integer,

details=true | false,

knotMax=double,

knotMethod

equal=integer

list={double-1 <, double-2, ...>}

listWithBoundary={double-1 <, double-2, ...>}

multiscale

endScale=integer

startScale=integer

}

rangeFractions={double-1 <, double-2, ...>}

knotMin=double,

name="string",

naturalCubic=true | false,

separate=true | false,

split=true | false,

vars={"variable-name-1" <, "variable-name-2", ...>}

}, {...}},

caslib="string",

compress=true | false,

indexVars={"variable-name-1" <, "variable-name-2", ...>},

label="string",

lifetime=64-bit-integer,

maxMemSize=64-bit-integer,

memoryFormat="DVR" | "INHERIT" | "STANDARD",

name="table-name",

onDemand=true | false,

promote=true | false,

replace=true | false,

replication=integer,

threadBlockSize=64-bit-integer,

timeStamp="string",

where={"string-1" <, "string-2", ...>}

storeCovariance=true | false,

caslib="string",

computedOnDemand=true | false,

computedVars

={{

format="string",

formattedLength=integer,

label="string",

name="variable-name",

nfd=integer,

nfl=integer

}, {...}},

computedVarsProgram="string",

dataSourceOptions={key-1=any-list-or-data-type-1 <, key-2=any-list-or-data-type-2, ...>},

groupBy

={{

format="string",

formattedLength=integer,

label="string",

name="variable-name",

nfd=integer,

nfl=integer

}, {...}},

groupByMode="NOSORT" | "REDISTRIBUTE",

name="table-name",

onDemand=true | false,

orderBy

={{

format="string",

formattedLength=integer,

label="string",

name="variable-name",

nfd=integer,

nfl=integer

}, {...}},

singlePass=true | false,

vars

={{

format="string",

formattedLength=integer,

label="string",

name="variable-name",

nfd=integer,

nfl=integer

}, {...}},

where="where-expression",

whereTable

casLib="string"

name="table-name"

vars

={{

format="string",

formattedLength=integer,

label="string",

name="variable-name",

nfd=integer,

nfl=integer

}, {...}}

where="where-expression"

}

weight="variable-name"

}

indicates a required parameter

Summary: Input and Output Tables

If a row includes a subparameter, you can specify the name, caslib, and so on in the subparameter. Otherwise, you can specify the name, caslib, and so on in the parameter.

Parameters for Reading Input Tables
Parameter	Subparameter	Description
required parameterfuncDef	—	specifies the name of the table that contains the FCMP function definitions.
inest	—	specifies a table to read initial parameter estimates from.
required parametertable	—	specifies the input data table.

Parameters for Creating Output Tables
Parameter	Subparameter	Description
outModelInfo	—	specifies an output table that contains the information about each candidate distribution.
outScoreLibrary	outById	specifies whether and how to create the FCMP code for scoring functions.
outest	required parametertable	specifies whether and how to create a table to write final parameter estimates to.
output	required parametercasOut	specifies the details of the output data table to write scores and quantiles to.
outputTables	names	lists the names of results tables to save as CAS tables on the server.
outstat	—	specifies an output table that contains the values of statistics of fit for each model whose parameter estimation process converges.
store	—	specifies a table that stores the model fit information.

Parameter Descriptions

balanceThreads=true | false

Default	true

class={{classStatement-1} <, {classStatement-2}, ...>}

specifies the classification variables.

For more information about specifying the class parameter, see the common classStatement parameter (Appendix A: Common Parameters).

Aliases	classVars
Aliases	nominal

classGlobalOpts={classopts}

specifies options to control levelization of the classification variables.

For more information about specifying the classGlobalOpts parameter, see the common classopts parameter (Appendix A: Common Parameters).

classLevelsPrint=true | false

when set to False, suppresses the display of class levels.

Default	true

collectionEffect={{collection-1} <, {collection-2}, ...>}

defines a set of variables that are treated as a single effect that has multiple degrees of freedom.

Alias	collection

The collection value can be one or more of the following:

details=true | false

when set to True, requests a table that shows additional details that are related to this effect.

Default	false

* name="string"

specifies the name of the effect.

* vars={"variable-name-1" <, "variable-name-2", ...>}

criterion="AD" | "AIC" | "AICC" | "CVM" | "KS" | "LOGLIK" | "SBC"

specifies the criterion to mark the best distribution in the distribution selection table.

Default	LOGLIK

Anderson-Darling statistic

AIC

Akaike's information criterion

AICC

corrected Akaike's information criterion

CVM

Cramer-von Mises statistic

KS

Kolmogorov-Smirnov statistic

LOGLIK

-2 log likelihood

SBC

Schwarz's Bayesian information criterion

Alias	BIC

ctLimits={{ctlim-1} <, {ctlim-2}, ...>}

specifies the global censoring and truncation limits.

The ctlim value can be one or more of the following:

* limit=double

global limit or threshold

* role="LC" | "LT" | "RC" | "RT"

role of the limit

LC

left-censoring

LT

left-truncation

RC

right-censoring

RT

right-truncation

dfMixture={dfmix}

specifies the parameters for computing representative estimates of the cumulative distribution function (CDF) that are used to assess a scale regression model.

Long form	dfMixture={method="FULL" \| "MEAN" \| "QUANTILE" \| "RANDOM"}
Shortcut form	dfMixture="FULL" \| "MEAN" \| "QUANTILE" \| "RANDOM"

The dfmix value can be one or more of the following:

meanType="EXPXBETA" | "XBETA"

specifies how to compute the mean of the scale parameter for the MEAN mixture method.

Default	XBETA

EXPXBETA

computes the scale parameter as mean(exp(log(S_k))), where S_k is the scale parameter of observation k.

XBETA

computes the scale parameter as exp(mean(log(S_k))), where S_k is the scale parameter of observation k.

method="FULL" | "MEAN" | "QUANTILE" | "RANDOM"

specifies which distributions to use in forming the representative mixture distribution.

Default	MEAN

FULL

uses as many distributions as the number of observations.

MEAN

uses one distribution with scale equal to the mean of the scales of all observations.

QUANTILE

uses K-1 distributions each with scale equal to one of the K-1 quantiles of the scales of all observations.

RANDOM

uses K distributions each with scale equal to the scale of one of K randomly chosen observations.

nQuantile=integer

specifies the number of quantiles to use for the QUANTILE mixture method.

Default	2
Minimum value	2

nRandom=integer

specifies the number of points to use for the RANDOM mixture method.

Default	15
Minimum value	1

seed=double

specifies the seed to use for the RANDOM mixture method.

display={displayTables}

specifies a list of results tables to send to the client for display.

For more information about specifying the display parameter, see the common displayTables parameter (Appendix A: Common Parameters).

distributions={"string-1" <, "string-2", ...>}

specifies the list of distribution names to analyze.

empiricalCDF={edfparms}

specifies the parameters for computing the empirical distribution function.

Long form	empiricalCDF={method="AUTO" \| "KAPLANMEIER" \| "MODIFIEDKM" \| "NOTURNBULL" \| "STANDARD" \| "TURNBULL"}
Shortcut form	empiricalCDF="AUTO" \| "KAPLANMEIER" \| "MODIFIEDKM" \| "NOTURNBULL" \| "STANDARD" \| "TURNBULL"

The edfparms value can be one or more of the following:

method="AUTO" | "KAPLANMEIER" | "MODIFIEDKM" | "NOTURNBULL" | "STANDARD" | "TURNBULL"

specifies the method to use for computing the empirical distribution function.

Default	AUTO

AUTO

automatically selects the best method based on data.

KAPLANMEIER

uses the Kaplan-Meier estimation method.

MODIFIEDKM

uses a modified Kaplan-Meier estimation method that ignores contributions from observations with very small risk sets.

NOTURNBULL

automatically selects the best non-Turnbull method. This essentially disables Turnbull's method, and interval-censored observations are approximated by the midpoint of the interval.

STANDARD

uses a standard empirical CDF that ignores censoring and truncation.

TURNBULL

uses Turnbull's method. This is used only when both left-censoring and right-censoring are specified.

riskSetLowerBound=double

specifies the lower bound on risk set size. This applies only to the modified Kaplan-Meier method.

Alias	rslb

riskSetLowerBoundAlpha=double

specifies the value to use for Alpha to compute the lower bound on the risk set size as C*(N**Alpha). This applies only to the modified Kaplan-Meier method.

Alias	alpha
Default	0.5
Range	(0, 1)

riskSetLowerBoundC=double

specifies the value to use for C to compute the lower bound on the risk set size as C*(N**Alpha). This applies only to the modified Kaplan-Meier method.

Alias	c
Default	1
Minimum value (exclusive)	0

turnbullEnsureMLE=true | false

Alias	ensureMLE
Default	false

turnbullMaxError=double

specifies the maximum relative error to be allowed between estimates of two consecutive iterations. This applies only to the Turnbull method.

Alias	eps
Default	1E-08

turnbullMaxIter=integer

specifies the maximum number of iterations to attempt to find the empirical estimates. This applies only to the Turnbull method.

Alias	maxiter
Default	500

turnbullZeroProb=double

Alias	zeroprob
Default	1E-08

estimation={nlopts}

specifies parameters that control various aspects of the parameter estimation process.

Alias	nloptions

Long form	estimation={technique="CONGRA" \| "DBLDOG" \| "DUQUAN" \| "NEWRAP" \| "NMSIMP" \| "NONE" \| "NRRIDG" \| "QUANEW" \| "TRUREG"}
Shortcut form	estimation="CONGRA" \| "DBLDOG" \| "DUQUAN" \| "NEWRAP" \| "NMSIMP" \| "NONE" \| "NRRIDG" \| "QUANEW" \| "TRUREG"

The nlopts value can be one or more of the following:

absConv=double

specifies an absolute function convergence criterion.

Alias	absTol
Minimum value (exclusive)	0

absFconv=double

specifies an absolute function difference convergence criterion.

Alias	absFtol
Minimum value (exclusive)	0

absGconv=double

specifies an absolute gradient convergence criterion.

Alias	absGtol
Minimum value (exclusive)	0

absXconv=double

specifies an absolute parameter convergence criterion.

Alias	absXtol
Minimum value (exclusive)	0

fConv=double

specifies a relative function convergence criterion.

Alias	fTol
Minimum value (exclusive)	0

fConv2=double

specifies a second function convergence criterion.

Alias	fTol2
Minimum value (exclusive)	0

fsize=double

specifies the FSIZE parameter of the relative function and relative gradient termination criteria.

Minimum value (exclusive)	0

gConv=double

specifies a relative gradient convergence criterion.

Alias	gTol
Minimum value (exclusive)	0

gConv2=double

specifies another relative gradient convergence criterion.

Alias	gTol2
Minimum value (exclusive)	0

maxFunc=double

specifies the maximum number of objective function evaluations in the optimization process.

Minimum value	0

maxIter=double

specifies the maximum number of iterations in the optimization process. The default varies by the technique.

Minimum value	0

maxTime=double

specifies an upper limit (in seconds) on the CPU time for the optimization process.

Minimum value (exclusive)	0

minIter=integer

specifies the minimum number of iterations in the optimization process.

Minimum value	0

restart=double

specifies the number of iterations after which the QUANEW or CONGRA technique is restarted with a steepest search direction.

Minimum value	1

singular=double

specifies the singularity criterion that is used for the inversion of the Hessian matrix.

Range	(0, 1)

technique="CONGRA" | "DBLDOG" | "DUQUAN" | "NEWRAP" | "NMSIMP" | "NONE" | "NRRIDG" | "QUANEW" | "TRUREG"

specifies the nonlinear optimization technique to use.

Default	TRUREG

CONGRA

conjugate-gradient

DBLDOG

double-dogleg

DUQUAN

dual quasi-Newton

NEWRAP

Newton-Raphson (line search and ridging)

NMSIMP

Nelder-Mead simplex

NONE

no optimization

NRRIDG

Newton-Raphson with ridging

QUANEW

quasi-Newton

TRUREG

trust region

techniqueSelect="CONGRA" | "DBLDOG" | "DUQUAN" | "NEWRAP" | "NMSIMP" | "NONE" | "NRRIDG" | "QUANEW" | "TRUREG"

specifies the nonlinear optimization technique to use for intermediate scale model selection steps.

CONGRA

conjugate-gradient

DBLDOG

double-dogleg

DUQUAN

dual quasi-Newton

NEWRAP

Newton-Raphson (line search and ridging)

NMSIMP

Nelder-Mead simplex

NONE

no optimization

NRRIDG

Newton-Raphson with ridging

QUANEW

quasi-Newton

TRUREG

trust region

varianceDivisor="DF" | "N"

specifies the denominator to use for computing the covariance estimates.

Alias	vardef
Default	DF

DF

uses degrees of freedom as the denominator.

N

uses the number of nonmissing observations as the denominator.

xConv=double

specifies the relative parameter convergence criterion.

Alias	xTol
Minimum value (exclusive)	0

xsize=double

specifies the XSIZE parameter r of the relative parameter termination criterion.

Minimum value (exclusive)	0

freq="variable-name"

specifies the observation frequency variable.

* funcDef={castable}

specifies the name of the table that contains the FCMP function definitions.

Long form	funcDef={name="table-name"}
Shortcut form	funcDef="table-name"

The castable value can be one or more of the following:

caslib="string"

specifies the caslib for the input table that you want to use with the action. By default, the active caslib is used. Specify a value only if you need to access a table from a different caslib.

dataSourceOptions={key-1=any-list-or-data-type-1 <, key-2=any-list-or-data-type-2, ...>}

specifies data source options.

Aliases	options
Aliases	dataSource

* name="table-name"

specifies the name of the input table.

whereTable={groupbytable}

The groupbytable value can be one or more of the following:

casLib="string"

specifies the caslib for the filter table. By default, the active caslib is used.

dataSourceOptions={adls_noreq-parameters | bigquery-parameters | cas_noreq-parameters | clouddex-parameters | db2-parameters | dnfs-parameters | esp-parameters | fedsvr-parameters | gcs_noreq-parameters | hadoop-parameters | hana-parameters | impala-parameters | jdbc-parameters | mongodb-parameters | mysql-parameters | odbc-parameters | oracle-parameters | path-parameters | postgres-parameters | redshift-parameters | s3-parameters | sapiq-parameters | sforce-parameters | snowflake-parameters | spark-parameters | spde-parameters | sqlserver-parameters | ss_noreq-parameters | teradata-parameters | vertica-parameters | yellowbrick-parameters}

specifies data source options.

Aliases	options
Aliases	dataSource

For more information about specifying the dataSourceOptions parameter, see the common dataSourceOptions parameter (Appendix A: Common Parameters).

importOptions={fileType="ANY" | "AUDIO" | "AUTO" | "BASESAS" | "CSV" | "DOCUMENT" | "DTA" | "ESP" | "EXCEL" | "FMT" | "HDAT" | "IMAGE" | "JMP" | "LASR" | "PARQUET" | "SPSS" | "VIDEO" | "XLS", fileType-specific-parameters}

specifies the settings for reading a table from a data source.

Alias	import

For more information about specifying the importOptions parameter, see the common importOptions parameter (Appendix A: Common Parameters).

* name="table-name"

specifies the name of the filter table.

vars={{casinvardesc-1} <, {casinvardesc-2}, ...>}

specifies the variable names to use from the filter table.

The casinvardesc value can be one or more of the following:

format="string"

specifies the format to apply to the variable.

formattedLength=integer

specifies the length of format field plus the length of the format precision.

label="string"

specifies the descriptive label for the variable.

* name="variable-name"

specifies the name for the variable.

nfd=integer

specifies the length of the format precision.

nfl=integer

specifies the length of the format field.

where="where-expression"

specifies an expression for subsetting the data from the filter table.

inest={castable}

specifies a table to read initial parameter estimates from.

Long form	inest={name="table-name"}
Shortcut form	inest="table-name"

The castable value can be one or more of the following:

caslib="string"

specifies the caslib for the input table that you want to use with the action. By default, the active caslib is used. Specify a value only if you need to access a table from a different caslib.

dataSourceOptions={key-1=any-list-or-data-type-1 <, key-2=any-list-or-data-type-2, ...>}

specifies data source options.

Aliases	options
Aliases	dataSource

groupBy={{casinvardesc-1} <, {casinvardesc-2}, ...>}

specifies the names of the variables to use for grouping results.

The casinvardesc value can be one or more of the following:

format="string"

specifies the format to apply to the variable.

formattedLength=integer

specifies the length of format field plus the length of the format precision.

label="string"

specifies the descriptive label for the variable.

* name="variable-name"

specifies the name for the variable.

nfd=integer

specifies the length of the format precision.

nfl=integer

specifies the length of the format field.

groupByMode="NOSORT" | "REDISTRIBUTE"

specifies how to create groups.

Default	NOSORT

NOSORT

group the data without sorting on each machine, and then groups the data again on the controller.

REDISTRIBUTE

transfers rows between nodes to guarantee ordering within groups. This method is slower.

importOptions={fileType="ANY" | "AUDIO" | "AUTO" | "BASESAS" | "CSV" | "DOCUMENT" | "DTA" | "ESP" | "EXCEL" | "FMT" | "HDAT" | "IMAGE" | "JMP" | "LASR" | "PARQUET" | "SPSS" | "VIDEO" | "XLS", fileType-specific-parameters}

specifies the settings for reading a table from a data source.

Alias	import

For more information about specifying the importOptions parameter, see the common importOptions parameter (Appendix A: Common Parameters).

* name="table-name"

specifies the name of the input table.

where="where-expression"

specifies an expression for subsetting the input data.

whereTable={groupbytable}

The groupbytable value can be one or more of the following:

casLib="string"

specifies the caslib for the filter table. By default, the active caslib is used.

dataSourceOptions={adls_noreq-parameters | bigquery-parameters | cas_noreq-parameters | clouddex-parameters | db2-parameters | dnfs-parameters | esp-parameters | fedsvr-parameters | gcs_noreq-parameters | hadoop-parameters | hana-parameters | impala-parameters | jdbc-parameters | mongodb-parameters | mysql-parameters | odbc-parameters | oracle-parameters | path-parameters | postgres-parameters | redshift-parameters | s3-parameters | sapiq-parameters | sforce-parameters | snowflake-parameters | spark-parameters | spde-parameters | sqlserver-parameters | ss_noreq-parameters | teradata-parameters | vertica-parameters | yellowbrick-parameters}

specifies data source options.

Aliases	options
Aliases	dataSource

For more information about specifying the dataSourceOptions parameter, see the common dataSourceOptions parameter (Appendix A: Common Parameters).

importOptions={fileType="ANY" | "AUDIO" | "AUTO" | "BASESAS" | "CSV" | "DOCUMENT" | "DTA" | "ESP" | "EXCEL" | "FMT" | "HDAT" | "IMAGE" | "JMP" | "LASR" | "PARQUET" | "SPSS" | "VIDEO" | "XLS", fileType-specific-parameters}

specifies the settings for reading a table from a data source.

Alias	import

For more information about specifying the importOptions parameter, see the common importOptions parameter (Appendix A: Common Parameters).

* name="table-name"

specifies the name of the filter table.

vars={{casinvardesc-1} <, {casinvardesc-2}, ...>}

specifies the variable names to use from the filter table.

The casinvardesc value can be one or more of the following:

format="string"

specifies the format to apply to the variable.

formattedLength=integer

specifies the length of format field plus the length of the format precision.

label="string"

specifies the descriptive label for the variable.

* name="variable-name"

specifies the name for the variable.

nfd=integer

specifies the length of the format precision.

nfl=integer

specifies the length of the format field.

where="where-expression"

specifies an expression for subsetting the data from the filter table.

initialDistributionParameters={{initdist-1} <, {initdist-2}, ...>}

specifies initial values of distribution parameters.

Alias	initVals

The initdist value can be one or more of the following:

* distribution="string"

specifies the distribution name.

Alias	dist

* parameters={{distparm-1} <, {distparm-2}, ...>}

specifies the list of name-value pairs for the parameters that you want to initialize.

Aliases	parm
Aliases	parms

The distparm value can be one or more of the following:

* name="string"

specifies the name of the distribution parameter that you want to initialize.

* value=double

specifies the initial value for the specified distribution parameter.

lossVariables={{lossrole-1} <, {lossrole-2}, ...>}

specifies variables related to the loss response, such as target, censoring, and truncation.

The lossrole value can be one or more of the following:

* name="variable-name"

specifies the name of the variable (must be present in the input table).

* role="LC" | "LT" | "RC" | "RT" | "TARGET"

specifies the role of the variable.

LC

specifies the left-censoring variable name.

LT

specifies the left-truncation variable name.

RC

specifies the right-censoring variable name.

RT

specifies the right-truncation variable name.

TARGET

specifies the loss (dependent or target) variable name.

multimemberEffect={{multimember-1} <, {multimember-2}, ...>}

For more information about specifying the multimemberEffect parameter, see the common multimember parameter (Appendix A: Common Parameters).

Aliases	multimember
Aliases	mmEffect

nClassLevelsPrint=integer

limits the display of class levels. The value 0 suppresses all levels.

Minimum value	0

noConstFitStats=true | false

when set to True, excludes any constant distribution parameters from the calculations of likelihood-based fit statistics.

Alias	noConstSOF
Default	false

objective={objSpec}

specifies a custom objective function. If you do not specify this parameter, the default objective function, which is the negative of the log likelihood, is used.

Alias	obj

The objSpec value can be one or more of the following:

* SASProgram="string"

specifies programming statements that compute the objective function. The computed value is minimized.

Aliases	SASCode
Aliases	code

* symbol="string"

specifies the objective function symbol.

Alias	sym

outDetailLevel=integer

specifies the level of detail to be printed to the output tables.

Default	1
Minimum value	0

outest={outest}

specifies whether and how to create a table to write final parameter estimates to.

The outest value can be one or more of the following:

covout=true | false

when set to True, writes covariance estimates to the OUTEST table.

Default	false

selectOut=true | false

when set to True, writes only the regression parameters that correspond to the selected effects to the OUTEST table.

Default	false

* table={casouttable}

specifies a table to write final parameter estimates to.

For more information about specifying the table parameter, see the common casouttable parameter (Appendix A: Common Parameters).

zeroEstForNotInModel=true | false

Alias	zeroEst
Default	false

outModelInfo={casouttable}

specifies an output table that contains the information about each candidate distribution.

For more information about specifying the outModelInfo parameter, see the common casouttable parameter (Appendix A: Common Parameters).

output={sevOutputStatement}

specifies the details of the output data table to write scores and quantiles to.

The sevOutputStatement value can be one or more of the following:

* casOut={casouttable}

specifies the settings for an output table.

For more information about specifying the casOut parameter, see the common casouttable parameter (Appendix A: Common Parameters).

copyVars="ALL" | "ALL_MODEL" | "ALL_NUMERIC" | {"variable-name-1" <, "variable-name-2", ...>}

quantiles={outquant}

specifies the quantiles to write to the output data table for each distribution.

The outquant value can be one or more of the following:

names={"string-1" <, "string-2", ...>}

specifies the names of the quantile variables. If this list of names is shorter than the list of CDF values, then quantile variables for the remaining CDF values get default names.

nDecimal=integer

specifies the numeric precision (number of digits after the decimal point) to use for creating default names of the quantile variables.

Minimum value	0

* points={{points-1} <, {points-2}, ...>}

specifies the CDF values to evaluate the quantile function at.

Alias	cdfValues

* p=double

Range	(0, 1)

scoreFunctions={{sfunc-1} <, {sfunc-2}, ...>}

specifies the scoring functions to evaluate and write to the output data table.

Alias	scores

The sfunc value can be one or more of the following:

arguments={double-1 <, double-2, ...>}

specifies a value for the first argument of the scoring function. If this value is not specified, then the response variable's value in the input data table is used.

Alias	arg

* name="string"

specifies the name of the scoring function. For each converged model based on distribution D, function D_<name> is evaluated and written as a variable in the output data table.

variable="string"

specifies the name of the column for the scoring function (default is the name of the scoring function itself).

Alias	var

outputTables={outputTables}

lists the names of results tables to save as CAS tables on the server.

For more information about specifying the outputTables parameter, see the common outputTables parameter (Appendix A: Common Parameters).

Alias	displayOut

outScoreLibrary={outscorelib}

specifies whether and how to create the FCMP code for scoring functions.

Alias	outscorelib

The outscorelib value can be one or more of the following:

commonPackage=true | false

when set to True, creates only one common package to contain all the scoring functions.

Alias	onePackage
Default	false

outById={casouttable}

specifies an output table that contains the unique identifier for each BY group. This is a required parameter when you request BY-group processing.

For more information about specifying the outById parameter, see the common casouttable parameter (Appendix A: Common Parameters).

outstat={casouttable}

specifies an output table that contains the values of statistics of fit for each model whose parameter estimation process converges.

For more information about specifying the outstat parameter, see the common casouttable parameter (Appendix A: Common Parameters).

plotTable={plotTable}

specifies the information to write to the PlotData results table. You can use this table to prepare plots.

Alias	plot

The plotTable value can be one or more of the following:

cdf=true | false

when set to True, includes the cumulative distribution function (CDF) information in the PlotData table.

Default	false

edfAlpha=double

specifies the confidence level to use for computing the confidence intervals for the empirical distribution function (EDF) estimates.

Default	0.05
Range	(0, 1)

maxObsPerModel=integer

Alias	maxObs
Default	5000
Minimum value	5

pdf=true | false

when set to True, includes the probability density function (PDF) information in the PlotData table.

Default	false

qq=true | false

when set to True, includes the quantile information in the PlotData table.

Default	false

polynomialEffect={{polynomial-1} <, {polynomial-2}, ...>}

For more information about specifying the polynomialEffect parameter, see the common polynomial parameter (Appendix A: Common Parameters).

Aliases	poly
Aliases	polynomial

probObserved=double

specifies the probability of observability to use with left-truncation specification.

Range	(0, 1)

sampleFraction=double

specifies the fraction of observations to use to form a random sample, which is used to compute initial parameter values and EDF-based fit statistics.

Alias	initFraction
Range	0–1

sampleSize=integer

specifies the approximate number of observations to use per node to form a random sample, which is used to compute initial parameter values and EDF-based fit statistics.

Alias	initSize
Default	10000
Minimum value	50

scaleModel={modelStatement}

specifies the effects to consider for the scale regression model.

The modelStatement value can be one or more of the following:

effects={{effect-1} <, {effect-2}, ...>}

The effect value can be one or more of the following:

interaction="BAR" | "CROSS" | "NONE"

specifies the type of interaction for the variables.

Alias	interact
Default	NONE

maxInteract=integer

eliminates interaction effects whose order is higher than the specified integer value when used in conjunction with the BAR interaction.

nest={"string-1" <, "string-2", ...>}

* vars={"string-1" <, "string-2", ...>}

specifies the variables to use in defining a term of the effect. You must specify at least one variable.

include=integer | {{effect-1} <, {effect-2}, ...>}

The effect value is specified as follows:

interaction="BAR" | "CROSS" | "NONE"

specifies the type of interaction for the variables.

Alias	interact
Default	NONE

maxInteract=integer

eliminates interaction effects whose order is higher than the specified integer value when used in conjunction with the BAR interaction.

nest={"string-1" <, "string-2", ...>}

* vars={"string-1" <, "string-2", ...>}

specifies the variables to use in defining a term of the effect. You must specify at least one variable.

informative=true | false

Default	false

offset="variable-name"

specifies a numeric offset variable. This variable cannot be a classification variable, a response variable, or one of the explanatory variables.

start=integer | {{effect-1} <, {effect-2}, ...>}

The effect value is specified as follows:

interaction="BAR" | "CROSS" | "NONE"

specifies the type of interaction for the variables.

Alias	interact
Default	NONE

maxInteract=integer

eliminates interaction effects whose order is higher than the specified integer value when used in conjunction with the BAR interaction.

nest={"string-1" <, "string-2", ...>}

* vars={"string-1" <, "string-2", ...>}

specifies the variables to use in defining a term of the effect. You must specify at least one variable.

seed=integer

specifies the seed for random selection of initialization sample.

Minimum value	0

selection={selectionStatement}

specifies scale model selection parameters.

Long form	selection={method="BACKWARD" \| "ELASTICNET" \| "FORWARD" \| "FORWARDSWAP" \| "NONE" \| "STEPWISE"}
Shortcut form	selection="BACKWARD" \| "ELASTICNET" \| "FORWARD" \| "FORWARDSWAP" \| "NONE" \| "STEPWISE"

The selectionStatement value can be one or more of the following:

candidates=integer | "ALL"

specifies the maximum number of candidates to display at each step of the selection process, when the detail level ALL is specified.

choose="AIC" | "AICC" | "DEFAULT" | "NONE" | "SBC"

specifies the criterion for choosing the model. The specified criterion is evaluated at each step of the selection process, and the model that yields the best value of the criterion is chosen.

competitive=true | false

Default	false

details="ALL" | "NONE" | "STEPS" | "SUMMARY"

specifies the level of detail to produce about the selection process.

Default	SUMMARY

elasticNetOptions={enOptions}

specifies options to use in performing elastic net selection methods.

The enOptions value can be one or more of the following:

absFConv=double

specifies the absolute function difference convergence criterion.

Alias	abstol
Default	1E-08
Minimum value	0

fConv=double

specifies the relative function difference convergence criterion.

Alias	fTol
Minimum value	0

gConv=double

specifies the relative gradient convergence criterion.

Alias	gTol
Minimum value	0

lambda={double-1 <, double-2, ...>}

specifies the regularization parameters in the elastic net selection method.

mixing={double-1 <, double-2, ...>}

specifies the elastic net mixing parameter.

numLambda=integer

specifies the number of regularization parameters in the elastic net selection method.

Alias	nLambda
Default	0
Minimum value	0

rho=double

specifies the scaling factor to use in computing minimum regularization parameter.

Range	(0, 1)

solver="ADMM" | "BFGS" | "LBFGS" | "NLP"

specifies a solver for elastic net selection.

enscale=true | false

when set to True, applies scaling to beta in the elastic net selection method.

Default	false

ensteps=integer

specifies the number of iterations to use in the elastic net selection method.

Default	50

hierarchy="DEFAULT" | "NONE" | "SINGLE" | "SINGLECLASS"

Default	DEFAULT

L2=double

specifies the L2 parameter in the elastic net selection method.

Default	0

L2HIGH=double

specifies the upper bound to use in searching for the L2 parameter in the elastic net selection method.

Alias	maxL2
Default	1

L2LOW=double

specifies the lower bound to use in searching for the L2 parameter in the elastic net selection method.

Alias	minL2
Default	0

maxEffects=integer

specifies the maximum number of effects in any model to consider during the selection process. This parameter is ignored for backward selection.

maxSteps=integer

specifies the maximum number of selection steps to perform.

method="BACKWARD" | "ELASTICNET" | "FORWARD" | "FORWARDSWAP" | "NONE" | "STEPWISE"

specifies the model selection method.

Default	STEPWISE

minEffects=integer

specifies the minimum number of effects in any model to consider during backward selection.

orderSelect=true | false

when set to True, shows effects in parameter estimates tables in the order in which they were added to the model.

Default	false

plots=true | false

when set to True, produces coefficientProgression and selectionSummaryForPlots tables that you can use to produce selection diagnostic plots.

Default	false

select="AIC" | "AICC" | "DEFAULT" | "SBC"

specifies the criterion to use in determining the order in which effects enter or leave at each step of the selection method. This parameter does not apply to LAR or LASSO selection.

stop="AIC" | "AICC" | "DEFAULT" | "NONE" | "SBC"

specifies a criterion that to use for stopping the selection process. If you do not specify a stop criterion, then the criterion that is used to select the model is also used as the stop criterion.

stopHorizon=integer

specifies the number of consecutive steps at which the stop criterion must worsen in order for a local extremum to be detected.

Default	3

selectionDistParmInit="FIRST" | "FULL" | "PER"

specifies the method of initializing the distribution parameters for the regression effect selection process.

Alias	slctDistInit
Default	FIRST

FIRST

initializes the distribution parameters once, based on the first candidate model.

FULL

initializes the distribution parameters once, based on the full model.

PER

reinitializes the distribution parameters for every candidate model.

splineEffect={{spline-1} <, {spline-2}, ...>}

expands variables into spline bases whose form depends on the specified parameters.

For more information about specifying the splineEffect parameter, see the common spline parameter (Appendix A: Common Parameters).

Alias	spline

store={casouttable}

specifies a table that stores the model fit information.

For more information about specifying the store parameter, see the common casouttable parameter (Appendix A: Common Parameters).

storeCovariance=true | false

when set to True, writes the parameter covariance estimates to the store that you specify in the store parameter.

Alias	storeCov
Default	true

* table={castable}

specifies the input data table.

For more information about specifying the table parameter, see the common castable parameter (Appendix A: Common Parameters).

weight="variable-name"

specifies the observation weight variable.

severity Action

Estimates parameters of the specified severity distribution models.

Python Syntax

Expand all Collapse all

results=s.severity.severity(

balanceThreads=True | False,

class

=[{

"countMissing":True | False,

"descending":True | False,

"ignoreMissing":True | False,

"levelizeRaw":True | False,

"maxLev":integer,

"order":"FORMATTED" | "FREQ" | "FREQFORMATTED" | "FREQINTERNAL" | "INTERNAL",

"ref":"FIRST" | "LAST" | double | "string",

"split":True | False,

"vars":["variable-name-1" <, "variable-name-2", ...>]

}<, {...}>],

classGlobalOpts

"countMissing":True | False,

"descending":True | False,

"ignoreMissing":True | False,

"levelizeRaw":True | False,

"maxLev":integer,

"order":"FORMATTED" | "FREQ" | "FREQFORMATTED" | "FREQINTERNAL" | "INTERNAL",

"ref":"FIRST" | "LAST" | double | "string",

"split":True | False

classLevelsPrint=True | False,

collectionEffect

=[{

"details":True | False,

"name":"string",

"vars":["variable-name-1" <, "variable-name-2", ...>]

}<, {...}>],

criterion="AD" | "AIC" | "AICC" | "CVM" | "KS" | "LOGLIK" | "SBC",

ctLimits

=[{

"limit":double,

"role":"LC" | "LT" | "RC" | "RT"

}<, {...}>],

dfMixture

"meanType":"EXPXBETA" | "XBETA",

"method":"FULL" | "MEAN" | "QUANTILE" | "RANDOM",

"nQuantile":integer,

"nRandom":integer,

"seed":double

display

"caseSensitive":True | False,

"exclude":True | False,

"excludeAll":True | False,

"keyIsPath":True | False,

"names":["string-1" <, "string-2", ...>],

"pathType":"LABEL" | "NAME",

"traceNames":True | False

distributions=["string-1" <, "string-2", ...>],

empiricalCDF

"riskSetLowerBound":double,

"riskSetLowerBoundAlpha":double,

"riskSetLowerBoundC":double,

"turnbullEnsureMLE":True | False,

"turnbullMaxError":double,

"turnbullMaxIter":integer,

"turnbullZeroProb":double

estimation

"absConv":double,

"absFconv":double,

"absGconv":double,

"absXconv":double,

"fConv":double,

"fConv2":double,

"fsize":double,

"gConv":double,

"gConv2":double,

"maxFunc":double,

"maxIter":double,

"maxTime":double,

"minIter":integer,

"restart":double,

"singular":double,

"varianceDivisor":"DF" | "N",

"xConv":double,

"xsize":double

freq="variable-name",

funcDef

"caslib":"string",

"dataSourceOptions":{"key-1":{any-list-or-data-type-1} <, "key-2":{any-list-or-data-type-2}, ...>},

"name":"table-name",

"whereTable"

"casLib":"string"

"dataSourceOptions":{adls_noreq-parameters | bigquery-parameters | cas_noreq-parameters | clouddex-parameters | db2-parameters | dnfs-parameters | esp-parameters | fedsvr-parameters | gcs_noreq-parameters | hadoop-parameters | hana-parameters | impala-parameters | jdbc-parameters | mongodb-parameters | mysql-parameters | odbc-parameters | oracle-parameters | path-parameters | postgres-parameters | redshift-parameters | s3-parameters | sapiq-parameters | sforce-parameters | snowflake-parameters | spark-parameters | spde-parameters | sqlserver-parameters | ss_noreq-parameters | teradata-parameters | vertica-parameters | yellowbrick-parameters}

"importOptions":{"fileType":"ANY" | "AUDIO" | "AUTO" | "BASESAS" | "CSV" | "DELIMITED" | "DOCUMENT" | "DTA" | "ESP" | "EXCEL" | "FMT" | "HDAT" | "IMAGE" | "JMP" | "LASR" | "PARQUET" | "SOUND" | "SPSS" | "VIDEO" | "XLS", fileType-specific-parameters}

"name":"table-name"

"vars"

:[{

"format":"string",

"formattedLength":integer,

"label":"string",

"name":"variable-name",

"nfd":integer,

"nfl":integer

}<, {...}>]

"where":"where-expression"

}

"caslib":"string",

"dataSourceOptions":{"key-1":{any-list-or-data-type-1} <, "key-2":{any-list-or-data-type-2}, ...>},

"groupBy"

:[{

"format":"string",

"formattedLength":integer,

"label":"string",

"name":"variable-name",

"nfd":integer,

"nfl":integer

}<, {...}>],

"groupByMode":"NOSORT" | "REDISTRIBUTE",

"name":"table-name",

"where":"where-expression",

"whereTable"

"casLib":"string"

"name":"table-name"

"vars"

:[{

"format":"string",

"formattedLength":integer,

"label":"string",

"name":"variable-name",

"nfd":integer,

"nfl":integer

}<, {...}>]

"where":"where-expression"

}

=[{

"distribution":"string",

"parameters"

:[{

"name":"string",

"value":double

}<, {...}>]

}<, {...}>],

=[{

"name":"variable-name",

"role":"LC" | "LT" | "RC" | "RT" | "TARGET"

}<, {...}>],

=[{

"details":True | False,

"name":"string",

"noEffect":True | False,

"stdize":True | False,

"vars":["variable-name-1" <, "variable-name-2", ...>],

"weight":["variable-name-1" <, "variable-name-2", ...>]

}<, {...}>],

nClassLevelsPrint=integer,

noConstFitStats=True | False,

"SASProgram":"string",

"symbol":"string"

outDetailLevel=integer,

"covout":True | False,

"selectOut":True | False,

"table"

"caslib":"string"

"compress":True | False

"indexVars":["variable-name-1" <, "variable-name-2", ...>]

"label":"string"

"lifetime":64-bit-integer

"maxMemSize":64-bit-integer

"memoryFormat":"DVR" | "INHERIT" | "STANDARD"

"name":"table-name"

"onDemand":True | False

"promote":True | False

"replace":True | False

"replication":integer

"threadBlockSize":64-bit-integer

"timeStamp":"string"

"where":["string-1" <, "string-2", ...>]

"zeroEstForNotInModel":True | False

"caslib":"string",

"compress":True | False,

"indexVars":["variable-name-1" <, "variable-name-2", ...>],

"label":"string",

"lifetime":64-bit-integer,

"maxMemSize":64-bit-integer,

"memoryFormat":"DVR" | "INHERIT" | "STANDARD",

"name":"table-name",

"onDemand":True | False,

"promote":True | False,

"replace":True | False,

"replication":integer,

"threadBlockSize":64-bit-integer,

"timeStamp":"string",

"where":["string-1" <, "string-2", ...>]

"casOut"

"caslib":"string"

"compress":True | False

"indexVars":["variable-name-1" <, "variable-name-2", ...>]

"label":"string"

"lifetime":64-bit-integer

"maxMemSize":64-bit-integer

"memoryFormat":"DVR" | "INHERIT" | "STANDARD"

"name":"table-name"

"onDemand":True | False

"promote":True | False

"replace":True | False

"replication":integer

"threadBlockSize":64-bit-integer

"timeStamp":"string"

"where":["string-1" <, "string-2", ...>]

"copyVars":"ALL" | "ALL_MODEL" | "ALL_NUMERIC" | ["variable-name-1" <, "variable-name-2", ...>],

"quantiles"

"names":["string-1" <, "string-2", ...>]

"nDecimal":integer

"points"

:[{

"p":double

}<, {...}>]

"scoreFunctions"

:[{

"arguments":[double-1 <, double-2, ...>],

"name":"string",

"variable":"string"

}<, {...}>]

"groupByVarsRaw":True | False,

"includeAll":True | False,

"names":["string-1" <, "string-2", ...>] | {"key-1":{casouttable-1} <, "key-2":{casouttable-2}, ...>},

"repeated":True | False,

"replace":True | False

"commonPackage":True | False,

"outById"

"caslib":"string"

"compress":True | False

"indexVars":["variable-name-1" <, "variable-name-2", ...>]

"label":"string"

"lifetime":64-bit-integer

"maxMemSize":64-bit-integer

"memoryFormat":"DVR" | "INHERIT" | "STANDARD"

"name":"table-name"

"onDemand":True | False

"promote":True | False

"replace":True | False

"replication":integer

"threadBlockSize":64-bit-integer

"timeStamp":"string"

"where":["string-1" <, "string-2", ...>]

}

"caslib":"string",

"compress":True | False,

"indexVars":["variable-name-1" <, "variable-name-2", ...>],

"label":"string",

"lifetime":64-bit-integer,

"maxMemSize":64-bit-integer,

"memoryFormat":"DVR" | "INHERIT" | "STANDARD",

"name":"table-name",

"onDemand":True | False,

"promote":True | False,

"replace":True | False,

"replication":integer,

"threadBlockSize":64-bit-integer,

"timeStamp":"string",

"where":["string-1" <, "string-2", ...>]

"cdf":True | False,

"edfAlpha":double,

"maxObsPerModel":integer,

"pdf":True | False,

"qq":True | False

=[{

"degree":integer,

"details":True | False,

"labelStyle"

"expand":True | False

"exponent":"string"

"includeName":True | False

"productSymbol":"NONE" | "string"

"mDegree":integer,

"name":"string",

"noSeparate":True | False,

"standardize"

"method":"MOMENTS" | "MRANGE" | "WMOMENTS"

"options":"CENTER" | "CENTERSCALE" | "NONE" | "SCALE"

"prefix":"NONE" | "string"

"vars":["variable-name-1" <, "variable-name-2", ...>]

}<, {...}>],

probObserved=double,

sampleFraction=double,

sampleSize=integer,

"effects"

:[{

"interaction":"BAR" | "CROSS" | "NONE",

"maxInteract":integer,

"nest":["string-1" <, "string-2", ...>],

"vars":["string-1" <, "string-2", ...>]

}<, {...}>],

"include":integer | [{effect-1} <, {effect-2}, ...>],

"informative":True | False,

"offset":"variable-name",

"start":integer | [{effect-1} <, {effect-2}, ...>]

seed=integer,

"candidates":integer | "ALL",

"choose":"AIC" | "AICC" | "DEFAULT" | "NONE" | "SBC",

"competitive":True | False,

"details":"ALL" | "NONE" | "STEPS" | "SUMMARY",

"elasticNetOptions"

"absFConv":double

"fConv":double

"gConv":double

"lambda_":[double-1 <, double-2, ...>]

"mixing":[double-1 <, double-2, ...>]

"numLambda":integer

"rho":double

"solver":"ADMM" | "BFGS" | "LBFGS" | "NLP"

"enscale":True | False,

"ensteps":integer,

"hierarchy":"DEFAULT" | "NONE" | "SINGLE" | "SINGLECLASS",

"L2":double,

"L2HIGH":double,

"L2LOW":double,

"maxEffects":integer,

"maxSteps":integer,

"minEffects":integer,

"orderSelect":True | False,

"plots":True | False,

"select":"AIC" | "AICC" | "DEFAULT" | "SBC",

"stop":"AIC" | "AICC" | "DEFAULT" | "NONE" | "SBC",

"stopHorizon":integer

=[{

"basis":"BSPLINE" | "TPF_DEFAULT" | "TPF_NOINT" | "TPF_NOINTANDNOPOWERS" | "TPF_NOPOWERS",

"dataBoundary":True | False,

"degree":integer,

"details":True | False,

"knotMax":double,

"knotMethod"

"equal":integer

"list":[double-1 <, double-2, ...>]

"listWithBoundary":[double-1 <, double-2, ...>]

"multiscale"

"endScale":integer

"startScale":integer

}

"rangeFractions":[double-1 <, double-2, ...>]

"knotMin":double,

"name":"string",

"naturalCubic":True | False,

"separate":True | False,

"split":True | False,

"vars":["variable-name-1" <, "variable-name-2", ...>]

}<, {...}>],

"caslib":"string",

"compress":True | False,

"indexVars":["variable-name-1" <, "variable-name-2", ...>],

"label":"string",

"lifetime":64-bit-integer,

"maxMemSize":64-bit-integer,

"memoryFormat":"DVR" | "INHERIT" | "STANDARD",

"name":"table-name",

"onDemand":True | False,

"promote":True | False,

"replace":True | False,

"replication":integer,

"threadBlockSize":64-bit-integer,

"timeStamp":"string",

"where":["string-1" <, "string-2", ...>]

storeCovariance=True | False,

"caslib":"string",

"computedOnDemand":True | False,

"computedVars"

:[{

"format":"string",

"formattedLength":integer,

"label":"string",

"name":"variable-name",

"nfd":integer,

"nfl":integer

}<, {...}>],

"computedVarsProgram":"string",

"dataSourceOptions":{"key-1":{any-list-or-data-type-1} <, "key-2":{any-list-or-data-type-2}, ...>},

"groupBy"

:[{

"format":"string",

"formattedLength":integer,

"label":"string",

"name":"variable-name",

"nfd":integer,

"nfl":integer

}<, {...}>],

"groupByMode":"NOSORT" | "REDISTRIBUTE",

"name":"table-name",

"onDemand":True | False,

"orderBy"

:[{

"format":"string",

"formattedLength":integer,

"label":"string",

"name":"variable-name",

"nfd":integer,

"nfl":integer

}<, {...}>],

"singlePass":True | False,

"vars"

:[{

"format":"string",

"formattedLength":integer,

"label":"string",

"name":"variable-name",

"nfd":integer,

"nfl":integer

}<, {...}>],

"where":"where-expression",

"whereTable"

"casLib":"string"

"name":"table-name"

"vars"

:[{

"format":"string",

"formattedLength":integer,

"label":"string",

"name":"variable-name",

"nfd":integer,

"nfl":integer

}<, {...}>]

"where":"where-expression"

}

weight="variable-name"

)

indicates a required parameter

Summary: Input and Output Tables

If a row includes a subparameter, you can specify the name, caslib, and so on in the subparameter. Otherwise, you can specify the name, caslib, and so on in the parameter.

Parameters for Reading Input Tables
Parameter	Subparameter	Description
required parameterfuncDef	—	specifies the name of the table that contains the FCMP function definitions.
inest	—	specifies a table to read initial parameter estimates from.
required parametertable	—	specifies the input data table.

Parameters for Creating Output Tables
Parameter	Subparameter	Description
outModelInfo	—	specifies an output table that contains the information about each candidate distribution.
outScoreLibrary	outById	specifies whether and how to create the FCMP code for scoring functions.
outest	required parametertable	specifies whether and how to create a table to write final parameter estimates to.
output	required parametercasOut	specifies the details of the output data table to write scores and quantiles to.
outputTables	names	lists the names of results tables to save as CAS tables on the server.
outstat	—	specifies an output table that contains the values of statistics of fit for each model whose parameter estimation process converges.
store	—	specifies a table that stores the model fit information.

Parameter Descriptions

balanceThreads=True | False

Default	True

class_=[{classStatement-1} <, {classStatement-2}, ...>]

specifies the classification variables.

For more information about specifying the class parameter, see the common classStatement parameter (Appendix A: Common Parameters).

Aliases	classVars
Aliases	nominal

classGlobalOpts={classopts}

specifies options to control levelization of the classification variables.

For more information about specifying the classGlobalOpts parameter, see the common classopts parameter (Appendix A: Common Parameters).

classLevelsPrint=True | False

when set to False, suppresses the display of class levels.

Default	True

collectionEffect=[{collection-1} <, {collection-2}, ...>]

defines a set of variables that are treated as a single effect that has multiple degrees of freedom.

Alias	collection

The collection value can be one or more of the following:

"details":True | False

when set to True, requests a table that shows additional details that are related to this effect.

Default	False

* "name":"string"

specifies the name of the effect.

* "vars":["variable-name-1" <, "variable-name-2", ...>]

criterion="AD" | "AIC" | "AICC" | "CVM" | "KS" | "LOGLIK" | "SBC"

specifies the criterion to mark the best distribution in the distribution selection table.

Default	LOGLIK

Anderson-Darling statistic

AIC

Akaike's information criterion

AICC

corrected Akaike's information criterion

CVM

Cramer-von Mises statistic

KS

Kolmogorov-Smirnov statistic

LOGLIK

-2 log likelihood

SBC

Schwarz's Bayesian information criterion

Alias	BIC

ctLimits=[{ctlim-1} <, {ctlim-2}, ...>]

specifies the global censoring and truncation limits.

The ctlim value can be one or more of the following:

* "limit":double

global limit or threshold

* "role":"LC" | "LT" | "RC" | "RT"

role of the limit

LC

left-censoring

LT

left-truncation

RC

right-censoring

RT

right-truncation

dfMixture={dfmix}

specifies the parameters for computing representative estimates of the cumulative distribution function (CDF) that are used to assess a scale regression model.

Long form	dfMixture={"method":"FULL" \| "MEAN" \| "QUANTILE" \| "RANDOM"}
Shortcut form	dfMixture="FULL" \| "MEAN" \| "QUANTILE" \| "RANDOM"

The dfmix value can be one or more of the following:

"meanType":"EXPXBETA" | "XBETA"

specifies how to compute the mean of the scale parameter for the MEAN mixture method.

Default	XBETA

EXPXBETA

computes the scale parameter as mean(exp(log(S_k))), where S_k is the scale parameter of observation k.

XBETA

computes the scale parameter as exp(mean(log(S_k))), where S_k is the scale parameter of observation k.

"method":"FULL" | "MEAN" | "QUANTILE" | "RANDOM"

specifies which distributions to use in forming the representative mixture distribution.

Default	MEAN

FULL

uses as many distributions as the number of observations.

MEAN

uses one distribution with scale equal to the mean of the scales of all observations.

QUANTILE

uses K-1 distributions each with scale equal to one of the K-1 quantiles of the scales of all observations.

RANDOM

uses K distributions each with scale equal to the scale of one of K randomly chosen observations.

"nQuantile":integer

specifies the number of quantiles to use for the QUANTILE mixture method.

Default	2
Minimum value	2

"nRandom":integer

specifies the number of points to use for the RANDOM mixture method.

Default	15
Minimum value	1

"seed":double

specifies the seed to use for the RANDOM mixture method.

display={displayTables}

specifies a list of results tables to send to the client for display.

For more information about specifying the display parameter, see the common displayTables parameter (Appendix A: Common Parameters).

distributions=["string-1" <, "string-2", ...>]

specifies the list of distribution names to analyze.

empiricalCDF={edfparms}

specifies the parameters for computing the empirical distribution function.

Long form	empiricalCDF={"method":"AUTO" \| "KAPLANMEIER" \| "MODIFIEDKM" \| "NOTURNBULL" \| "STANDARD" \| "TURNBULL"}
Shortcut form	empiricalCDF="AUTO" \| "KAPLANMEIER" \| "MODIFIEDKM" \| "NOTURNBULL" \| "STANDARD" \| "TURNBULL"

The edfparms value can be one or more of the following:

"method":"AUTO" | "KAPLANMEIER" | "MODIFIEDKM" | "NOTURNBULL" | "STANDARD" | "TURNBULL"

specifies the method to use for computing the empirical distribution function.

Default	AUTO

AUTO

automatically selects the best method based on data.

KAPLANMEIER

uses the Kaplan-Meier estimation method.

MODIFIEDKM

uses a modified Kaplan-Meier estimation method that ignores contributions from observations with very small risk sets.

NOTURNBULL

automatically selects the best non-Turnbull method. This essentially disables Turnbull's method, and interval-censored observations are approximated by the midpoint of the interval.

STANDARD

uses a standard empirical CDF that ignores censoring and truncation.

TURNBULL

uses Turnbull's method. This is used only when both left-censoring and right-censoring are specified.

"riskSetLowerBound":double

specifies the lower bound on risk set size. This applies only to the modified Kaplan-Meier method.

Alias	rslb

"riskSetLowerBoundAlpha":double

specifies the value to use for Alpha to compute the lower bound on the risk set size as C*(N**Alpha). This applies only to the modified Kaplan-Meier method.

Alias	alpha
Default	0.5
Range	(0, 1)

"riskSetLowerBoundC":double

specifies the value to use for C to compute the lower bound on the risk set size as C*(N**Alpha). This applies only to the modified Kaplan-Meier method.

Alias	c
Default	1
Minimum value (exclusive)	0

"turnbullEnsureMLE":True | False

Alias	ensureMLE
Default	False

"turnbullMaxError":double

specifies the maximum relative error to be allowed between estimates of two consecutive iterations. This applies only to the Turnbull method.

Alias	eps
Default	1E-08

"turnbullMaxIter":integer

specifies the maximum number of iterations to attempt to find the empirical estimates. This applies only to the Turnbull method.

Alias	maxiter
Default	500

"turnbullZeroProb":double

Alias	zeroprob
Default	1E-08

estimation={nlopts}

specifies parameters that control various aspects of the parameter estimation process.

Alias	nloptions

Long form	estimation={"technique":"CONGRA" \| "DBLDOG" \| "DUQUAN" \| "NEWRAP" \| "NMSIMP" \| "NONE" \| "NRRIDG" \| "QUANEW" \| "TRUREG"}
Shortcut form	estimation="CONGRA" \| "DBLDOG" \| "DUQUAN" \| "NEWRAP" \| "NMSIMP" \| "NONE" \| "NRRIDG" \| "QUANEW" \| "TRUREG"

The nlopts value can be one or more of the following:

"absConv":double

specifies an absolute function convergence criterion.

Alias	absTol
Minimum value (exclusive)	0

"absFconv":double

specifies an absolute function difference convergence criterion.

Alias	absFtol
Minimum value (exclusive)	0

"absGconv":double

specifies an absolute gradient convergence criterion.

Alias	absGtol
Minimum value (exclusive)	0

"absXconv":double

specifies an absolute parameter convergence criterion.

Alias	absXtol
Minimum value (exclusive)	0

"fConv":double

specifies a relative function convergence criterion.

Alias	fTol
Minimum value (exclusive)	0

"fConv2":double

specifies a second function convergence criterion.

Alias	fTol2
Minimum value (exclusive)	0

"fsize":double

specifies the FSIZE parameter of the relative function and relative gradient termination criteria.

Minimum value (exclusive)	0

"gConv":double

specifies a relative gradient convergence criterion.

Alias	gTol
Minimum value (exclusive)	0

"gConv2":double

specifies another relative gradient convergence criterion.

Alias	gTol2
Minimum value (exclusive)	0

"maxFunc":double

specifies the maximum number of objective function evaluations in the optimization process.

Minimum value	0

"maxIter":double

specifies the maximum number of iterations in the optimization process. The default varies by the technique.

Minimum value	0

"maxTime":double

specifies an upper limit (in seconds) on the CPU time for the optimization process.

Minimum value (exclusive)	0

"minIter":integer

specifies the minimum number of iterations in the optimization process.

Minimum value	0

"restart":double

specifies the number of iterations after which the QUANEW or CONGRA technique is restarted with a steepest search direction.

Minimum value	1

"singular":double

specifies the singularity criterion that is used for the inversion of the Hessian matrix.

Range	(0, 1)

"technique":"CONGRA" | "DBLDOG" | "DUQUAN" | "NEWRAP" | "NMSIMP" | "NONE" | "NRRIDG" | "QUANEW" | "TRUREG"

specifies the nonlinear optimization technique to use.

Default	TRUREG

CONGRA

conjugate-gradient

DBLDOG

double-dogleg

DUQUAN

dual quasi-Newton

NEWRAP

Newton-Raphson (line search and ridging)

NMSIMP

Nelder-Mead simplex

NONE

no optimization

NRRIDG

Newton-Raphson with ridging

QUANEW

quasi-Newton

TRUREG

trust region

"techniqueSelect":"CONGRA" | "DBLDOG" | "DUQUAN" | "NEWRAP" | "NMSIMP" | "NONE" | "NRRIDG" | "QUANEW" | "TRUREG"

specifies the nonlinear optimization technique to use for intermediate scale model selection steps.

CONGRA

conjugate-gradient

DBLDOG

double-dogleg

DUQUAN

dual quasi-Newton

NEWRAP

Newton-Raphson (line search and ridging)

NMSIMP

Nelder-Mead simplex

NONE

no optimization

NRRIDG

Newton-Raphson with ridging

QUANEW

quasi-Newton

TRUREG

trust region

"varianceDivisor":"DF" | "N"

specifies the denominator to use for computing the covariance estimates.

Alias	vardef
Default	DF

DF

uses degrees of freedom as the denominator.

N

uses the number of nonmissing observations as the denominator.

"xConv":double

specifies the relative parameter convergence criterion.

Alias	xTol
Minimum value (exclusive)	0

"xsize":double

specifies the XSIZE parameter r of the relative parameter termination criterion.

Minimum value (exclusive)	0

freq="variable-name"

specifies the observation frequency variable.

* funcDef={castable}

specifies the name of the table that contains the FCMP function definitions.

Long form	funcDef={"name":"table-name"}
Shortcut form	funcDef="table-name"

The castable value can be one or more of the following:

"caslib":"string"

specifies the caslib for the input table that you want to use with the action. By default, the active caslib is used. Specify a value only if you need to access a table from a different caslib.

"dataSourceOptions":{"key-1":{any-list-or-data-type-1} <, "key-2":{any-list-or-data-type-2}, ...>}

specifies data source options.

Aliases	options
Aliases	dataSource

* "name":"table-name"

specifies the name of the input table.

"whereTable":{groupbytable}

The groupbytable value can be one or more of the following:

"casLib":"string"

specifies the caslib for the filter table. By default, the active caslib is used.

"dataSourceOptions":{adls_noreq-parameters | bigquery-parameters | cas_noreq-parameters | clouddex-parameters | db2-parameters | dnfs-parameters | esp-parameters | fedsvr-parameters | gcs_noreq-parameters | hadoop-parameters | hana-parameters | impala-parameters | jdbc-parameters | mongodb-parameters | mysql-parameters | odbc-parameters | oracle-parameters | path-parameters | postgres-parameters | redshift-parameters | s3-parameters | sapiq-parameters | sforce-parameters | snowflake-parameters | spark-parameters | spde-parameters | sqlserver-parameters | ss_noreq-parameters | teradata-parameters | vertica-parameters | yellowbrick-parameters}

specifies data source options.

Aliases	options
Aliases	dataSource

For more information about specifying the dataSourceOptions parameter, see the common dataSourceOptions parameter (Appendix A: Common Parameters).

"importOptions":{"fileType":"ANY" | "AUDIO" | "AUTO" | "BASESAS" | "CSV" | "DOCUMENT" | "DTA" | "ESP" | "EXCEL" | "FMT" | "HDAT" | "IMAGE" | "JMP" | "LASR" | "PARQUET" | "SPSS" | "VIDEO" | "XLS", fileType-specific-parameters}

specifies the settings for reading a table from a data source.

Alias	import_

For more information about specifying the importOptions parameter, see the common importOptions parameter (Appendix A: Common Parameters).

* "name":"table-name"

specifies the name of the filter table.

"vars":[{casinvardesc-1} <, {casinvardesc-2}, ...>]

specifies the variable names to use from the filter table.

The casinvardesc value can be one or more of the following:

"format":"string"

specifies the format to apply to the variable.

"formattedLength":integer

specifies the length of format field plus the length of the format precision.

"label":"string"

specifies the descriptive label for the variable.

* "name":"variable-name"

specifies the name for the variable.

"nfd":integer

specifies the length of the format precision.

"nfl":integer

specifies the length of the format field.

"where":"where-expression"

specifies an expression for subsetting the data from the filter table.

inest={castable}

specifies a table to read initial parameter estimates from.

Long form	inest={"name":"table-name"}
Shortcut form	inest="table-name"

The castable value can be one or more of the following:

"caslib":"string"

specifies the caslib for the input table that you want to use with the action. By default, the active caslib is used. Specify a value only if you need to access a table from a different caslib.

"dataSourceOptions":{"key-1":{any-list-or-data-type-1} <, "key-2":{any-list-or-data-type-2}, ...>}

specifies data source options.

Aliases	options
Aliases	dataSource

"groupBy":[{casinvardesc-1} <, {casinvardesc-2}, ...>]

specifies the names of the variables to use for grouping results.

The casinvardesc value can be one or more of the following:

"format":"string"

specifies the format to apply to the variable.

"formattedLength":integer

specifies the length of format field plus the length of the format precision.

"label":"string"

specifies the descriptive label for the variable.

* "name":"variable-name"

specifies the name for the variable.

"nfd":integer

specifies the length of the format precision.

"nfl":integer

specifies the length of the format field.

"groupByMode":"NOSORT" | "REDISTRIBUTE"

specifies how to create groups.

Default	NOSORT

NOSORT

group the data without sorting on each machine, and then groups the data again on the controller.

REDISTRIBUTE

transfers rows between nodes to guarantee ordering within groups. This method is slower.

"importOptions":{"fileType":"ANY" | "AUDIO" | "AUTO" | "BASESAS" | "CSV" | "DOCUMENT" | "DTA" | "ESP" | "EXCEL" | "FMT" | "HDAT" | "IMAGE" | "JMP" | "LASR" | "PARQUET" | "SPSS" | "VIDEO" | "XLS", fileType-specific-parameters}

specifies the settings for reading a table from a data source.

Alias	import_

For more information about specifying the importOptions parameter, see the common importOptions parameter (Appendix A: Common Parameters).

* "name":"table-name"

specifies the name of the input table.

"where":"where-expression"

specifies an expression for subsetting the input data.

"whereTable":{groupbytable}

The groupbytable value can be one or more of the following:

"casLib":"string"

specifies the caslib for the filter table. By default, the active caslib is used.

"dataSourceOptions":{adls_noreq-parameters | bigquery-parameters | cas_noreq-parameters | clouddex-parameters | db2-parameters | dnfs-parameters | esp-parameters | fedsvr-parameters | gcs_noreq-parameters | hadoop-parameters | hana-parameters | impala-parameters | jdbc-parameters | mongodb-parameters | mysql-parameters | odbc-parameters | oracle-parameters | path-parameters | postgres-parameters | redshift-parameters | s3-parameters | sapiq-parameters | sforce-parameters | snowflake-parameters | spark-parameters | spde-parameters | sqlserver-parameters | ss_noreq-parameters | teradata-parameters | vertica-parameters | yellowbrick-parameters}

specifies data source options.

Aliases	options
Aliases	dataSource

For more information about specifying the dataSourceOptions parameter, see the common dataSourceOptions parameter (Appendix A: Common Parameters).

"importOptions":{"fileType":"ANY" | "AUDIO" | "AUTO" | "BASESAS" | "CSV" | "DOCUMENT" | "DTA" | "ESP" | "EXCEL" | "FMT" | "HDAT" | "IMAGE" | "JMP" | "LASR" | "PARQUET" | "SPSS" | "VIDEO" | "XLS", fileType-specific-parameters}

specifies the settings for reading a table from a data source.

Alias	import_

For more information about specifying the importOptions parameter, see the common importOptions parameter (Appendix A: Common Parameters).

* "name":"table-name"

specifies the name of the filter table.

"vars":[{casinvardesc-1} <, {casinvardesc-2}, ...>]

specifies the variable names to use from the filter table.

The casinvardesc value can be one or more of the following:

"format":"string"

specifies the format to apply to the variable.

"formattedLength":integer

specifies the length of format field plus the length of the format precision.

"label":"string"

specifies the descriptive label for the variable.

* "name":"variable-name"

specifies the name for the variable.

"nfd":integer

specifies the length of the format precision.

"nfl":integer

specifies the length of the format field.

"where":"where-expression"

specifies an expression for subsetting the data from the filter table.

initialDistributionParameters=[{initdist-1} <, {initdist-2}, ...>]

specifies initial values of distribution parameters.

Alias	initVals

The initdist value can be one or more of the following:

* "distribution":"string"

specifies the distribution name.

Alias	dist

* "parameters":[{distparm-1} <, {distparm-2}, ...>]

specifies the list of name-value pairs for the parameters that you want to initialize.

Aliases	parm
Aliases	parms

The distparm value can be one or more of the following:

* "name":"string"

specifies the name of the distribution parameter that you want to initialize.

* "value":double

specifies the initial value for the specified distribution parameter.

lossVariables=[{lossrole-1} <, {lossrole-2}, ...>]

specifies variables related to the loss response, such as target, censoring, and truncation.

The lossrole value can be one or more of the following:

* "name":"variable-name"

specifies the name of the variable (must be present in the input table).

* "role":"LC" | "LT" | "RC" | "RT" | "TARGET"

specifies the role of the variable.

LC

specifies the left-censoring variable name.

LT

specifies the left-truncation variable name.

RC

specifies the right-censoring variable name.

RT

specifies the right-truncation variable name.

TARGET

specifies the loss (dependent or target) variable name.

multimemberEffect=[{multimember-1} <, {multimember-2}, ...>]

For more information about specifying the multimemberEffect parameter, see the common multimember parameter (Appendix A: Common Parameters).

Aliases	multimember
Aliases	mmEffect

nClassLevelsPrint=integer

limits the display of class levels. The value 0 suppresses all levels.

Minimum value	0

noConstFitStats=True | False

when set to True, excludes any constant distribution parameters from the calculations of likelihood-based fit statistics.

Alias	noConstSOF
Default	False

objective={objSpec}

specifies a custom objective function. If you do not specify this parameter, the default objective function, which is the negative of the log likelihood, is used.

Alias	obj

The objSpec value can be one or more of the following:

* "SASProgram":"string"

specifies programming statements that compute the objective function. The computed value is minimized.

Aliases	SASCode
Aliases	code

* "symbol":"string"

specifies the objective function symbol.

Alias	sym

outDetailLevel=integer

specifies the level of detail to be printed to the output tables.

Default	1
Minimum value	0

outest={outest}

specifies whether and how to create a table to write final parameter estimates to.

The outest value can be one or more of the following:

"covout":True | False

when set to True, writes covariance estimates to the OUTEST table.

Default	False

"selectOut":True | False

when set to True, writes only the regression parameters that correspond to the selected effects to the OUTEST table.

Default	False

* "table":{casouttable}

specifies a table to write final parameter estimates to.

For more information about specifying the table parameter, see the common casouttable parameter (Appendix A: Common Parameters).

"zeroEstForNotInModel":True | False

Alias	zeroEst
Default	False

outModelInfo={casouttable}

specifies an output table that contains the information about each candidate distribution.

For more information about specifying the outModelInfo parameter, see the common casouttable parameter (Appendix A: Common Parameters).

output={sevOutputStatement}

specifies the details of the output data table to write scores and quantiles to.

The sevOutputStatement value can be one or more of the following:

* "casOut":{casouttable}

specifies the settings for an output table.

For more information about specifying the casOut parameter, see the common casouttable parameter (Appendix A: Common Parameters).

"copyVars":"ALL" | "ALL_MODEL" | "ALL_NUMERIC" | ["variable-name-1" <, "variable-name-2", ...>]

"quantiles":{outquant}

specifies the quantiles to write to the output data table for each distribution.

The outquant value can be one or more of the following:

"names":["string-1" <, "string-2", ...>]

specifies the names of the quantile variables. If this list of names is shorter than the list of CDF values, then quantile variables for the remaining CDF values get default names.

"nDecimal":integer

specifies the numeric precision (number of digits after the decimal point) to use for creating default names of the quantile variables.

Minimum value	0

* "points":[{points-1} <, {points-2}, ...>]

specifies the CDF values to evaluate the quantile function at.

Alias	cdfValues

* "p":double

Range	(0, 1)

"scoreFunctions":[{sfunc-1} <, {sfunc-2}, ...>]

specifies the scoring functions to evaluate and write to the output data table.

Alias	scores

The sfunc value can be one or more of the following:

"arguments":[double-1 <, double-2, ...>]

specifies a value for the first argument of the scoring function. If this value is not specified, then the response variable's value in the input data table is used.

Alias	arg

* "name":"string"

specifies the name of the scoring function. For each converged model based on distribution D, function D_<name> is evaluated and written as a variable in the output data table.

"variable":"string"

specifies the name of the column for the scoring function (default is the name of the scoring function itself).

Alias	var

outputTables={outputTables}

lists the names of results tables to save as CAS tables on the server.

For more information about specifying the outputTables parameter, see the common outputTables parameter (Appendix A: Common Parameters).

Alias	displayOut

outScoreLibrary={outscorelib}

specifies whether and how to create the FCMP code for scoring functions.

Alias	outscorelib

The outscorelib value can be one or more of the following:

"commonPackage":True | False

when set to True, creates only one common package to contain all the scoring functions.

Alias	onePackage
Default	False

"outById":{casouttable}

specifies an output table that contains the unique identifier for each BY group. This is a required parameter when you request BY-group processing.

For more information about specifying the outById parameter, see the common casouttable parameter (Appendix A: Common Parameters).

outstat={casouttable}

specifies an output table that contains the values of statistics of fit for each model whose parameter estimation process converges.

For more information about specifying the outstat parameter, see the common casouttable parameter (Appendix A: Common Parameters).

plotTable={plotTable}

specifies the information to write to the PlotData results table. You can use this table to prepare plots.

Alias	plot

The plotTable value can be one or more of the following:

"cdf":True | False

when set to True, includes the cumulative distribution function (CDF) information in the PlotData table.

Default	False

"edfAlpha":double

specifies the confidence level to use for computing the confidence intervals for the empirical distribution function (EDF) estimates.

Default	0.05
Range	(0, 1)

"maxObsPerModel":integer

Alias	maxObs
Default	5000
Minimum value	5

"pdf":True | False

when set to True, includes the probability density function (PDF) information in the PlotData table.

Default	False

"qq":True | False

when set to True, includes the quantile information in the PlotData table.

Default	False

polynomialEffect=[{polynomial-1} <, {polynomial-2}, ...>]

For more information about specifying the polynomialEffect parameter, see the common polynomial parameter (Appendix A: Common Parameters).

Aliases	poly
Aliases	polynomial

probObserved=double

specifies the probability of observability to use with left-truncation specification.

Range	(0, 1)

sampleFraction=double

specifies the fraction of observations to use to form a random sample, which is used to compute initial parameter values and EDF-based fit statistics.

Alias	initFraction
Range	0–1

sampleSize=integer

specifies the approximate number of observations to use per node to form a random sample, which is used to compute initial parameter values and EDF-based fit statistics.

Alias	initSize
Default	10000
Minimum value	50

scaleModel={modelStatement}

specifies the effects to consider for the scale regression model.

The modelStatement value can be one or more of the following:

"effects":[{effect-1} <, {effect-2}, ...>]

The effect value can be one or more of the following:

"interaction":"BAR" | "CROSS" | "NONE"

specifies the type of interaction for the variables.

Alias	interact
Default	NONE

"maxInteract":integer

eliminates interaction effects whose order is higher than the specified integer value when used in conjunction with the BAR interaction.

"nest":["string-1" <, "string-2", ...>]

* "vars":["string-1" <, "string-2", ...>]

specifies the variables to use in defining a term of the effect. You must specify at least one variable.

"include":integer | [{effect-1} <, {effect-2}, ...>]

The effect value is specified as follows:

"interaction":"BAR" | "CROSS" | "NONE"

specifies the type of interaction for the variables.

Alias	interact
Default	NONE

"maxInteract":integer

eliminates interaction effects whose order is higher than the specified integer value when used in conjunction with the BAR interaction.

"nest":["string-1" <, "string-2", ...>]

* "vars":["string-1" <, "string-2", ...>]

specifies the variables to use in defining a term of the effect. You must specify at least one variable.

"informative":True | False

Default	False

"offset":"variable-name"

specifies a numeric offset variable. This variable cannot be a classification variable, a response variable, or one of the explanatory variables.

"start":integer | [{effect-1} <, {effect-2}, ...>]

The effect value is specified as follows:

"interaction":"BAR" | "CROSS" | "NONE"

specifies the type of interaction for the variables.

Alias	interact
Default	NONE

"maxInteract":integer

eliminates interaction effects whose order is higher than the specified integer value when used in conjunction with the BAR interaction.

"nest":["string-1" <, "string-2", ...>]

* "vars":["string-1" <, "string-2", ...>]

specifies the variables to use in defining a term of the effect. You must specify at least one variable.

seed=integer

specifies the seed for random selection of initialization sample.

Minimum value	0

selection={selectionStatement}

specifies scale model selection parameters.

Long form	selection={"method":"BACKWARD" \| "ELASTICNET" \| "FORWARD" \| "FORWARDSWAP" \| "NONE" \| "STEPWISE"}
Shortcut form	selection="BACKWARD" \| "ELASTICNET" \| "FORWARD" \| "FORWARDSWAP" \| "NONE" \| "STEPWISE"

The selectionStatement value can be one or more of the following:

"candidates":integer | "ALL"

specifies the maximum number of candidates to display at each step of the selection process, when the detail level ALL is specified.

"choose":"AIC" | "AICC" | "DEFAULT" | "NONE" | "SBC"

specifies the criterion for choosing the model. The specified criterion is evaluated at each step of the selection process, and the model that yields the best value of the criterion is chosen.

"competitive":True | False

Default	False

"details":"ALL" | "NONE" | "STEPS" | "SUMMARY"

specifies the level of detail to produce about the selection process.

Default	SUMMARY

"elasticNetOptions":{enOptions}

specifies options to use in performing elastic net selection methods.

The enOptions value can be one or more of the following:

"absFConv":double

specifies the absolute function difference convergence criterion.

Alias	abstol
Default	1E-08
Minimum value	0

"fConv":double

specifies the relative function difference convergence criterion.

Alias	fTol
Minimum value	0

"gConv":double

specifies the relative gradient convergence criterion.

Alias	gTol
Minimum value	0

"lambda_":[double-1 <, double-2, ...>]

specifies the regularization parameters in the elastic net selection method.

"mixing":[double-1 <, double-2, ...>]

specifies the elastic net mixing parameter.

"numLambda":integer

specifies the number of regularization parameters in the elastic net selection method.

Alias	nLambda
Default	0
Minimum value	0

"rho":double

specifies the scaling factor to use in computing minimum regularization parameter.

Range	(0, 1)

"solver":"ADMM" | "BFGS" | "LBFGS" | "NLP"

specifies a solver for elastic net selection.

"enscale":True | False

when set to True, applies scaling to beta in the elastic net selection method.

Default	False

"ensteps":integer

specifies the number of iterations to use in the elastic net selection method.

Default	50

"hierarchy":"DEFAULT" | "NONE" | "SINGLE" | "SINGLECLASS"

Default	DEFAULT

"L2":double

specifies the L2 parameter in the elastic net selection method.

Default	0

"L2HIGH":double

specifies the upper bound to use in searching for the L2 parameter in the elastic net selection method.

Alias	maxL2
Default	1

"L2LOW":double

specifies the lower bound to use in searching for the L2 parameter in the elastic net selection method.

Alias	minL2
Default	0

"maxEffects":integer

specifies the maximum number of effects in any model to consider during the selection process. This parameter is ignored for backward selection.

"maxSteps":integer

specifies the maximum number of selection steps to perform.

"method":"BACKWARD" | "ELASTICNET" | "FORWARD" | "FORWARDSWAP" | "NONE" | "STEPWISE"

specifies the model selection method.

Default	STEPWISE

"minEffects":integer

specifies the minimum number of effects in any model to consider during backward selection.

"orderSelect":True | False

when set to True, shows effects in parameter estimates tables in the order in which they were added to the model.

Default	False

"plots":True | False

when set to True, produces coefficientProgression and selectionSummaryForPlots tables that you can use to produce selection diagnostic plots.

Default	False

"select":"AIC" | "AICC" | "DEFAULT" | "SBC"

specifies the criterion to use in determining the order in which effects enter or leave at each step of the selection method. This parameter does not apply to LAR or LASSO selection.

"stop":"AIC" | "AICC" | "DEFAULT" | "NONE" | "SBC"

specifies a criterion that to use for stopping the selection process. If you do not specify a stop criterion, then the criterion that is used to select the model is also used as the stop criterion.

"stopHorizon":integer

specifies the number of consecutive steps at which the stop criterion must worsen in order for a local extremum to be detected.

Default	3

selectionDistParmInit="FIRST" | "FULL" | "PER"

specifies the method of initializing the distribution parameters for the regression effect selection process.

Alias	slctDistInit
Default	FIRST

FIRST

initializes the distribution parameters once, based on the first candidate model.

FULL

initializes the distribution parameters once, based on the full model.

PER

reinitializes the distribution parameters for every candidate model.

splineEffect=[{spline-1} <, {spline-2}, ...>]

expands variables into spline bases whose form depends on the specified parameters.

For more information about specifying the splineEffect parameter, see the common spline parameter (Appendix A: Common Parameters).

Alias	spline

store={casouttable}

specifies a table that stores the model fit information.

For more information about specifying the store parameter, see the common casouttable parameter (Appendix A: Common Parameters).

storeCovariance=True | False

when set to True, writes the parameter covariance estimates to the store that you specify in the store parameter.

Alias	storeCov
Default	True

* table={castable}

specifies the input data table.

For more information about specifying the table parameter, see the common castable parameter (Appendix A: Common Parameters).

weight="variable-name"

specifies the observation weight variable.

severity Action

Estimates parameters of the specified severity distribution models.

R Syntax

Expand all Collapse all

results <– cas.severity.severity(s,

balanceThreads=TRUE | FALSE,

class

=list( list(

countMissing=TRUE | FALSE,

descending=TRUE | FALSE,

ignoreMissing=TRUE | FALSE,

levelizeRaw=TRUE | FALSE,

maxLev=integer,

order="FORMATTED" | "FREQ" | "FREQFORMATTED" | "FREQINTERNAL" | "INTERNAL",

ref="FIRST" | "LAST" | double | "string",

split=TRUE | FALSE,

vars=list("variable-name-1" <, "variable-name-2", ...>)

) <, list(...)>),

classGlobalOpts

=list(

countMissing=TRUE | FALSE,

descending=TRUE | FALSE,

ignoreMissing=TRUE | FALSE,

levelizeRaw=TRUE | FALSE,

maxLev=integer,

order="FORMATTED" | "FREQ" | "FREQFORMATTED" | "FREQINTERNAL" | "INTERNAL",

ref="FIRST" | "LAST" | double | "string",

split=TRUE | FALSE

classLevelsPrint=TRUE | FALSE,

collectionEffect

=list( list(

details=TRUE | FALSE,

name="string",

vars=list("variable-name-1" <, "variable-name-2", ...>)

) <, list(...)>),

criterion="AD" | "AIC" | "AICC" | "CVM" | "KS" | "LOGLIK" | "SBC",

ctLimits

=list( list(

limit=double,

role="LC" | "LT" | "RC" | "RT"

) <, list(...)>),

dfMixture

=list(

meanType="EXPXBETA" | "XBETA",

method="FULL" | "MEAN" | "QUANTILE" | "RANDOM",

nQuantile=integer,

nRandom=integer,

seed=double

display

=list(

caseSensitive=TRUE | FALSE,

exclude=TRUE | FALSE,

excludeAll=TRUE | FALSE,

keyIsPath=TRUE | FALSE,

names=list("string-1" <, "string-2", ...>),

pathType="LABEL" | "NAME",

traceNames=TRUE | FALSE

distributions=list("string-1" <, "string-2", ...>),

empiricalCDF

=list(

riskSetLowerBound=double,

riskSetLowerBoundAlpha=double,

riskSetLowerBoundC=double,

turnbullEnsureMLE=TRUE | FALSE,

turnbullMaxError=double,

turnbullMaxIter=integer,

turnbullZeroProb=double

estimation

=list(

absConv=double,

absFconv=double,

absGconv=double,

absXconv=double,

fConv=double,

fConv2=double,

fsize=double,

gConv=double,

gConv2=double,

maxFunc=double,

maxIter=double,

maxTime=double,

minIter=integer,

restart=double,

singular=double,

varianceDivisor="DF" | "N",

xConv=double,

xsize=double

freq="variable-name",

funcDef

=list(

caslib="string",

dataSourceOptions=list(key-1=list(any-list-or-data-type-1) <, key-2=list(any-list-or-data-type-2), ...>),

name="table-name",

=list(

casLib="string"

name="table-name"

=list( list(

format="string",

formattedLength=integer,

label="string",

name="variable-name",

nfd=integer,

nfl=integer

) <, list(...)>)

where="where-expression"

)

=list(

caslib="string",

dataSourceOptions=list(key-1=list(any-list-or-data-type-1) <, key-2=list(any-list-or-data-type-2), ...>),

groupBy

=list( list(

format="string",

formattedLength=integer,

label="string",

name="variable-name",

nfd=integer,

nfl=integer

) <, list(...)>),

groupByMode="NOSORT" | "REDISTRIBUTE",

name="table-name",

where="where-expression",

=list(

casLib="string"

name="table-name"

=list( list(

format="string",

formattedLength=integer,

label="string",

name="variable-name",

nfd=integer,

nfl=integer

) <, list(...)>)

where="where-expression"

)

=list( list(

distribution="string",

parameters

=list( list(

name="string",

value=double

) <, list(...)>)

) <, list(...)>),

=list( list(

name="variable-name",

role="LC" | "LT" | "RC" | "RT" | "TARGET"

) <, list(...)>),

=list( list(

details=TRUE | FALSE,

name="string",

noEffect=TRUE | FALSE,

stdize=TRUE | FALSE,

vars=list("variable-name-1" <, "variable-name-2", ...>),

weight=list("variable-name-1" <, "variable-name-2", ...>)

) <, list(...)>),

nClassLevelsPrint=integer,

noConstFitStats=TRUE | FALSE,

=list(

SASProgram="string",

symbol="string"

outDetailLevel=integer,

=list(

covout=TRUE | FALSE,

selectOut=TRUE | FALSE,

=list(

caslib="string"

compress=TRUE | FALSE

indexVars=list("variable-name-1" <, "variable-name-2", ...>)

label="string"

lifetime=64-bit-integer

maxMemSize=64-bit-integer

memoryFormat="DVR" | "INHERIT" | "STANDARD"

name="table-name"

onDemand=TRUE | FALSE

promote=TRUE | FALSE

replace=TRUE | FALSE

replication=integer

threadBlockSize=64-bit-integer

timeStamp="string"

where=list("string-1" <, "string-2", ...>)

zeroEstForNotInModel=TRUE | FALSE

=list(

caslib="string",

compress=TRUE | FALSE,

indexVars=list("variable-name-1" <, "variable-name-2", ...>),

label="string",

lifetime=64-bit-integer,

maxMemSize=64-bit-integer,

memoryFormat="DVR" | "INHERIT" | "STANDARD",

name="table-name",

onDemand=TRUE | FALSE,

promote=TRUE | FALSE,

replace=TRUE | FALSE,

replication=integer,

threadBlockSize=64-bit-integer,

timeStamp="string",

where=list("string-1" <, "string-2", ...>)

=list(

casOut

=list(

caslib="string"

compress=TRUE | FALSE

indexVars=list("variable-name-1" <, "variable-name-2", ...>)

label="string"

lifetime=64-bit-integer

maxMemSize=64-bit-integer

memoryFormat="DVR" | "INHERIT" | "STANDARD"

name="table-name"

onDemand=TRUE | FALSE

promote=TRUE | FALSE

replace=TRUE | FALSE

replication=integer

threadBlockSize=64-bit-integer

timeStamp="string"

where=list("string-1" <, "string-2", ...>)

copyVars="ALL" | "ALL_MODEL" | "ALL_NUMERIC" | list("variable-name-1" <, "variable-name-2", ...>),

quantiles

=list(

names=list("string-1" <, "string-2", ...>)

nDecimal=integer

points

=list( list(

p=double

) <, list(...)>)

scoreFunctions

=list( list(

arguments=list(double-1 <, double-2, ...>),

name="string",

variable="string"

) <, list(...)>)

=list(

groupByVarsRaw=TRUE | FALSE,

includeAll=TRUE | FALSE,

names=list("string-1" <, "string-2", ...>) | list(key-1=list(casouttable-1) <, key-2=list(casouttable-2), ...>),

repeated=TRUE | FALSE,

replace=TRUE | FALSE

=list(

commonPackage=TRUE | FALSE,

outById

=list(

caslib="string"

compress=TRUE | FALSE

indexVars=list("variable-name-1" <, "variable-name-2", ...>)

label="string"

lifetime=64-bit-integer

maxMemSize=64-bit-integer

memoryFormat="DVR" | "INHERIT" | "STANDARD"

name="table-name"

onDemand=TRUE | FALSE

promote=TRUE | FALSE

replace=TRUE | FALSE

replication=integer

threadBlockSize=64-bit-integer

timeStamp="string"

where=list("string-1" <, "string-2", ...>)

)

=list(

caslib="string",

compress=TRUE | FALSE,

indexVars=list("variable-name-1" <, "variable-name-2", ...>),

label="string",

lifetime=64-bit-integer,

maxMemSize=64-bit-integer,

memoryFormat="DVR" | "INHERIT" | "STANDARD",

name="table-name",

onDemand=TRUE | FALSE,

promote=TRUE | FALSE,

replace=TRUE | FALSE,

replication=integer,

threadBlockSize=64-bit-integer,

timeStamp="string",

where=list("string-1" <, "string-2", ...>)

=list(

cdf=TRUE | FALSE,

edfAlpha=double,

maxObsPerModel=integer,

pdf=TRUE | FALSE,

qq=TRUE | FALSE

=list( list(

degree=integer,

details=TRUE | FALSE,

labelStyle

=list(

expand=TRUE | FALSE

exponent="string"

includeName=TRUE | FALSE

productSymbol="NONE" | "string"

mDegree=integer,

name="string",

noSeparate=TRUE | FALSE,

standardize

=list(

method="MOMENTS" | "MRANGE" | "WMOMENTS"

options="CENTER" | "CENTERSCALE" | "NONE" | "SCALE"

prefix="NONE" | "string"

vars=list("variable-name-1" <, "variable-name-2", ...>)

) <, list(...)>),

probObserved=double,

sampleFraction=double,

sampleSize=integer,

=list(

effects

=list( list(

interaction="BAR" | "CROSS" | "NONE",

maxInteract=integer,

nest=list("string-1" <, "string-2", ...>),

vars=list("string-1" <, "string-2", ...>)

) <, list(...)>),

include=integer | list( list(effect-1) <, list(effect-2), ...>),

informative=TRUE | FALSE,

offset="variable-name",

start=integer | list( list(effect-1) <, list(effect-2), ...>)

seed=integer,

=list(

candidates=integer | "ALL",

choose="AIC" | "AICC" | "DEFAULT" | "NONE" | "SBC",

competitive=TRUE | FALSE,

details="ALL" | "NONE" | "STEPS" | "SUMMARY",

elasticNetOptions

=list(

absFConv=double

fConv=double

gConv=double

lambda=list(double-1 <, double-2, ...>)

mixing=list(double-1 <, double-2, ...>)

numLambda=integer

rho=double

solver="ADMM" | "BFGS" | "LBFGS" | "NLP"

enscale=TRUE | FALSE,

ensteps=integer,

hierarchy="DEFAULT" | "NONE" | "SINGLE" | "SINGLECLASS",

L2=double,

L2HIGH=double,

L2LOW=double,

maxEffects=integer,

maxSteps=integer,

minEffects=integer,

orderSelect=TRUE | FALSE,

plots=TRUE | FALSE,

select="AIC" | "AICC" | "DEFAULT" | "SBC",

stop="AIC" | "AICC" | "DEFAULT" | "NONE" | "SBC",

stopHorizon=integer

=list( list(

basis="BSPLINE" | "TPF_DEFAULT" | "TPF_NOINT" | "TPF_NOINTANDNOPOWERS" | "TPF_NOPOWERS",

dataBoundary=TRUE | FALSE,

degree=integer,

details=TRUE | FALSE,

knotMax=double,

knotMethod

=list(

equal=integer

list=list(double-1 <, double-2, ...>)

listWithBoundary=list(double-1 <, double-2, ...>)

multiscale

=list(

endScale=integer

startScale=integer

)

rangeFractions=list(double-1 <, double-2, ...>)

knotMin=double,

name="string",

naturalCubic=TRUE | FALSE,

separate=TRUE | FALSE,

split=TRUE | FALSE,

vars=list("variable-name-1" <, "variable-name-2", ...>)

) <, list(...)>),

=list(

caslib="string",

compress=TRUE | FALSE,

indexVars=list("variable-name-1" <, "variable-name-2", ...>),

label="string",

lifetime=64-bit-integer,

maxMemSize=64-bit-integer,

memoryFormat="DVR" | "INHERIT" | "STANDARD",

name="table-name",

onDemand=TRUE | FALSE,

promote=TRUE | FALSE,

replace=TRUE | FALSE,

replication=integer,

threadBlockSize=64-bit-integer,

timeStamp="string",

where=list("string-1" <, "string-2", ...>)

storeCovariance=TRUE | FALSE,