Dave: Continuing with response to 3 (margins macro call).   ODS trace on; ODS output _Margins PERSIST=RUN ; %margins(version=1.02,data=SHE810SL.TOTALS,class=TCcd,response=TOTMCRCLMTS2, model=TCcd TOTMCRCLMTS1, dist=GAMMA, link=LOG,roptions=event='2', margins=TCcd TOTMCRCLMTS1, options=diff cl) %margins(version=1.02,data=SHE810SL.TOTALS,class=TCcd,response=TOTMCRCLMTS2, model=TCcd TOTMCRCLMTS1 MDCRcovTS1 MDCRcovTS2 HMOcovTS1 HMOcovTS2, dist=GAMMA, link=LOG, roptions=event='2', margins=TCcd TOTMCRCLMTS1 HMOcovTS1 HMOcovTS2, options=diff cl) ODS _all_ close;         ... View more

CCW VRDC: Release: 9.04.01M5PO91317 System: LIN X64 Linux   The macro does not, for example, create _Margins or _MEffect.   I'll have to download the program and copy the macro call. S         ... View more

Dave: I am trying to run the margins macro under SAS/EG within the CCW VRDC portal. While it generates standard GENMOD estimates, it doesn't appear to recognize the Margin= or Effect= parameters. I've suppressed the search for a newer version of the macro in a copy. That eliminated one error. The dist=gamma and link=log specifications do work.   (BTW, the VRDC requires approval for any download and doesn't allow copying of programs or logs directly. If you need to see them, I'll have to request approval to download.)   The log of the program is displaying another error message: Libref SUBFUNC is not assigned. I don't find a Libref in the macro. Is this part of the macro or something strange happening in the VRDC SAS/EG environment?   Also, is the reference value of a binary independent variable set the same in the macro as in GENMOD? Does the macro parameter apply only to CLASS variables?   Finally, how does the margins macro differ from the QLIM plots of marginals? A basic TOBIT model may give us an alternative for comparison. Thanks in advance, S ... View more

Alex: Learning to link data tables via hash indexes offers you a chance to learn more about how database systems optimize inner joins. I suspect that your professor knows that in SAS SQL an ordinary inner join of two tables, if a hash index of one will fit into memory, will optimize as a scan of a dynamic hash index. Even so, once you learn how to program the inner join using a hash index, it will be easier to learn how to program a left join using a hash index. The SAS SQL compiler does not, last time I checked, optimize a left join. So except for academic purposes, a SAS SQL SELECT ... FROM r1 INNER JOIN r2 ON r1.key = r2.key will usually execute as fast or faster than a Data step that generates a hash index. Over on the SAS-L side, that discussion has continued off and on for decades. S ... View more

Try changing this section of the SELECT statement ,                 ...                 left join                  ecroext.ext_mcf_filtered b                   on a.PERSON_KEY = b.PERSON_KEY        where b.SRVC_FROM_DT between "01Mar2016"d and "31Aug2017"d        and  b.Client_ky = 2;   quit ; ,      to                 ...                 left join                  (select * from ecroext.ext_mcf_filtered where SRVC_FROM_DT between "01Mar2016"d and "31Aug2017"d                                                                                   and  Client_ky = 2) b                   on a.PERSON_KEY = b.PERSON_KEY        ;   quit ;   The SAS SQL compiler ideally promotes a WHERE condition when equivalent as I suggest, Sometimes one must intervene and improve the query prior to compilation. If ecroext.ext_mcf_filtered contains a large number of column variables, you might also move the SELECT arguments down to the subquery and add the PERSON_KEY. (Remember to remove the column variable aliases in the subquery). S ... View more

Deduplication of rows in a very large SAS data set turns out to be easy to do with a hash object. Difficulties arise when deciding on how to deduplicate.    The DISTINCT operator in SAS SQL eliminates rows that have exactly the same column variable values. SORT NODUPKEY keeps on the first row among multiple rows with the same key value(s). The hash object  allows one to define a key of one or more column variables, look it up on a hash index as a Data step reads each row of a SAS Data step, and decide how to write or suppress key duplicates. As a safeguard, it often makes sense to write the rows with the first instance of a key value to a main data set and rows with subsequent values of that same key value to a data set of excluded duplicates.   How do you prefer to handle the duplicates? Note that if one is eliminating key duplicates, the duplicates may contain data not found in the row being kept. S     ... View more