The PHSELECT Procedure
PROC PHSELECT Statement
PROC PHSELECT <options>;
The PROC PHSELECT statement invokes the procedure. Table 1 summarizes the available options in the PROC PHSELECT statement by function. They are then described fully in alphabetical order.
Table 1: PROC PHSELECT Statement Options
| Option | Description |
|---|---|
| ALPHA= | Specifies a global significance level |
| DATA= | Specifies the CAS input data table |
| Output Options | |
| CORRB | Displays the "Parameter Estimates Correlation Matrix" table |
| COVB | Displays the "Parameter Estimates Covariance Matrix" table |
| ITHIST | Displays the "Iteration History" table |
| LOGLIKENULL | Displays the "–2 Log Likelihood for Null Model" table |
| NOCLPRINT | Limits or suppresses the display of class levels |
| NOSTDERR | Suppresses computation of the covariance matrix and standard errors |
| Optimization Options | |
| ABSCONV= | Tunes the absolute function convergence criterion |
| ABSFCONV= | Tunes the absolute function difference convergence criterion |
| ABSGCONV= | Tunes the absolute gradient convergence criterion |
| ABSXCONV= | Tunes the absolute parameter convergence criterion |
| FCONV= | Tunes the relative function difference convergence criterion |
| FCONV2= | Tunes the second relative function difference convergence criterion |
| GCONV= | Tunes the relative gradient convergence criterion |
| GCONV2= | Tunes the second relative gradient convergence criterion |
| XCONV= | Tunes the relative gradient convergence criterion |
| HESSIAN | Uses analytic Hessian instead of finite-difference Hessian |
| MAXFUNC= | Specifies the maximum number of function evaluations in any optimization |
| MAXITER= | Specifies the maximum number of iterations in any optimization |
| MAXTIME= | Specifies the upper limit of CPU time (in seconds) for any optimization |
| MINITER= | Specifies the minimum number of iterations in any optimization |
| TECHNIQUE= | Selects the optimization technique |
| LASSO Options | |
| LASSORHO= | Specifies the base regularization parameter for the LASSO method |
| LASSOSTEPS= | Specifies the maximum number of steps for the LASSO method |
| LASSOTOL= | Specifies the convergence criterion for the LASSO method |
The optimization options, with the exception of the HESSIAN option, are fully described in the section Optimization Options in Chapter 2, Shared Concepts. The following list describes the other options available in the PROC PHSELECT statement:
- ALPHA=number
specifies a global significance level for the construction of confidence intervals. The confidence level is 1–number. The value of number must be between 0 and 1. By default, ALPHA=0.05.
- CORRB
creates the "Parameter Estimates Correlation Matrix" table. The correlation matrix is computed by normalizing the covariance matrix
. That is, if 
is an element of , then the corresponding element of the correlation matrix is 
, where 
. - COVB
creates the "Parameter Estimates Covariance Matrix" table. The covariance matrix is computed as the inverse of the negative Hessian matrix, which is the matrix of second derivatives of the log-likelihood function with respect to the model parameters.
- DATA=CAS-libref.data-table
-
names the input data table for PROC PHSELECT to use. The default is the most recently created data table. CAS-libref.data-table is a two-level name, where
- CAS-libref
refers to a collection of information that is defined in the LIBNAME statement and includes the
caslib, which includes a path to the data, and a session identifier, which defaults to the active session but which can be explicitly defined in the LIBNAME statement. For more information about CAS-libref, see the section Using CAS Sessions and CAS Engine Librefs.- data-table
specifies the name of the input data table.
- HESSIAN
computes the Hessian matrix by using the analytic expression for the second-order derivatives of the log partial likelihood function instead of using the finite-difference method. When you specify this option, the optimization technique defaults to the Newton-Raphson method with ridging (TECH=NRRIDG), but you can use the TECH= option to specify the technique of your choice. The HESSIAN option requires a large amount of memory for each machine in the cluster.
- ITHIST
- LASSORHO=r
specifies the base regularization parameter for the LASSO model selection method. The regularization parameter for step i is r
. By default, LASSORHO=0.8. - LASSOSTEPS=n
specifies the maximum number of steps for LASSO model selection. By default, LASSOSTEPS=20.
- LASSOTOL=r
specifies the convergence tolerance for the optimization algorithm that solves for the LASSO parameter estimates at each step of LASSO model selection. By default, LASSOTOL=1E–6.
- LOGLIKENULL
creates the "–2 Log Likelihood for the Null Model" table. If you also specify the PARTITION statement, the table displays the –2 log likelihood for the null model for each data partition.
- NOCLPRINT<=number>
suppresses the display of the "Class Level Information" table if you do not specify number. If you specify number, the values of the classification variables are displayed for only those variables whose number of levels is less than number. Specifying number helps reduce the size of the "Class Level Information" table if some classification variables have a large number of levels.
- NOSTDERR
suppresses computation of the covariance matrix and the standard errors of the regression coefficients. When the model contains many variables (such as thousands), inverting the Hessian matrix to derive the covariance matrix and the standard errors of the regression coefficients can be time-consuming. The CORRB, COVB, and TYPE3 options are not available when you specify this option.