The PROC MWPCA statement invokes the procedure. Table 1 summarizes the options available in the PROC MWPCA statement.
Table 1: PROC MWPCA Statement Options
| Option | Description |
|---|
| Input Data Table Option |
|---|
| DATA= | Specifies the input data table |
| Other Options |
|---|
| CENTER | Centers the input data |
| CUMEIGPCTTOL= | Specifies the significance level of eigenvalues that determine the rank of the low-rank matrix in each window |
| NTHREADS= | Specifies the maximum number of threads to use on each computation node |
| ROBUST | Uses robust principal component analysis on each window |
| STEPSIZE= | Specifies the step size |
| WINDOWSIZE= | Specifies the window size |
You must specify the following options:
-
STEPSIZE=number
specifies the step size, which is the number of observations between the beginnings of two consecutive windows.
-
WINDOWSIZE=number
specifies the number of observations in each window.
You can also specify the following options:
-
CENTER
centers the observations by the mean of each column for each window.
-
CUMEIGPCTTOL=number
specifies the significance level of eigenvalues that determine the rank of the low-rank matrix in each window. This value must be between 0 and 1. By default, CUMEIGPCTTOL=1. For more information, see the section Cumulative Eigenvalue Percentage Tolerance.
-
DATA=CAS-libref.data-table
names the input data table for PROC MWPCA 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.
-
NTHREADS=number-of-threads
specifies the number of threads to use in the computation. The default value is the lesser of the number of threads available per computation node and 16.
-
ROBUST
uses the robust principal component analysis (RPCA) method to obtain principal components for each window. You can specify the RPCA options in the RPCA statement. For more information about the RPCA method, see Chapter 19, The RPCA Procedure. If you do not specify this option, the regular PCA method is used on each window. It is recommended not to use this option for anomaly detection.