ECM Procedure

MARGINAL Statement

MARGINAL marginal-variable: DATA=libref.data-table <marginal-options>;

MGL marginal-variable: DATA=libref.data-table <marginal-options>;

The MARGINAL statement enables you to specify information about one marginal-variable. Specify one MARGINAL statement for each marginal variable that you want to analyze.

You must specify the name of the marginal-variable followed by a colon (':') as the first item in the MARGINAL statement. This marginal-variable must appear in the copula simulation data table that you specify in the DATA= option in the PROC ECM statement.

Next, you must specify the data table by using the following option:

DATA=libref.data-table

specifies the data table that contains an empirical sample of the marginal-variable. libref.data-table is a two-level name, where libref refers to the library, and data-table specifies the name of the input data table. For more information about this two-level name, see the DATA= option and the section Using CAS Sessions and CAS Engine Librefs. The libref must be identical to the libref that you specify in the DATA= option in the PROC ECM statement.

Typically, you specify a data table that was generated by the CCDM procedure in order to simulate a compound distribution sample from the underlying frequency and severity models that the CNTSELECT and SEVSELECT procedures create.

You can specify the following marginal-options to specify the variable and observations that contain the marginal variable sample:

SAMPLEVAR=sample-variable SVAR=sample-variable

specifies the variable that contains the marginal sample in the marginal data table.

If the name of the sample-variable is the same as the marginal-variable, then you do not need to specify this option. If PROC ECM cannot find a variable named marginal-variable in the marginal table and if you do not specify this option, then the sample variable name is assumed to be _AGGSEV_, which must be present in the marginal data table.

DRAWID=sample-identifier SAMPLEID=sample-identifier

identifies the sample to use when the marginal data table contains multiple samples of the marginal variable, where sample-identifier is an integer.

The data table can contain multiple samples when it is a result of the perturbation analysis that the CCDM procedure conducts. Each sample is identified by the value of the _DRAWID_ variable. If you specify a sample-identifier value of k, then PROC ECM uses the observations from the marginal data table whose _DRAWID_=k.

If you do not specify either this option or the DRAWID= option in the PROC ECM statement, then PROC ECM uses a default value of 0 when the marginal table contains the _DRAWID_ variable.

The value that you specify for this option overrides the value that you specify for the DRAWID= option in the PROC ECM statement.

PCTLDEF=1 | 2 | 3 | 4 | 5

specifies the method of computing the percentiles. The value that you specify for this option overrides the value that you specify for the PCTLDEF= option in the PROC ECM statement.

You can specify the following values:

1: uses the weighted average.
2: uses the value closest to the sample size times the percentile.
3: uses the empirical distribution function.
4: uses the weighted average (identical to PCTLDEF=1).
5: uses the empirical distribution function with averaging.

For more information about each method, see the section Percentile Computation Methods.

By default, PCTLDEF=5.

Last updated: July 09, 2026