PROC QUANTSELECT saves the list of selected effects in a macro variable so that you can use other SAS procedures to perform post-selection analyses. This list does not explicitly include the intercept so that you can use it in the MODEL statement of other SAS/STAT regression procedures.
Table 84.12 describes the macro variables that PROC QUANTSELECT creates. When multiple quantile levels or BY processing are used, one macro variable, indexed by the quantile level order and the BY group number, is created for each quantile level and BY group combination.
Table 84.12: Macro Variables Created for Subsequent Processing
Macro Variable |
Description |
---|---|
Single Quantile Level and No BY processing |
|
|
Selected model |
Multiple Quantile Levels and No BY Processing |
|
|
Number of quantile levels |
|
Selected model for the first quantile level |
|
Selected model for the second quantile level |
… |
|
Single Quantile Level and BY Processing |
|
|
Number of BY groups |
|
Selected model for BY group 1 |
|
Selected model for BY group 2 |
… |
|
Multiple Quantile Levels and BY Processing |
|
|
Number of quantile levels |
|
Number of BY groups |
|
Selected model for the first quantile level and BY group 1 |
|
Selected model for the second quantile level and BY group 1 |
… |
|
|
Selected model for the first quantile level and BY group 2 |
|
Selected model for the second quantile level and BY group 2 |
… |
The macro variables _QRSIND
, _QRSINDT1
, _QRSIND1
, and _QRSIND1T1
are all synonyms. If you do not specify multiple quantile levels or BY processing, the macro variables _QRSNUMTAUS
and _QRSNUMBYS
are both set to 1.
PROC QUANTSELECT creates two output data set variables, _BY_
and _QUANTILE_
, to aid in associating macro variables with output data set observations when multiple quantile levels or BY processing are
used. The values of these two variables are integers that match the i,j components of the macro variable names _QRSINDiTj
.