Programming the statistical procedures from SAS

Odds ratio table in 9.3

Accepted Solution Solved
Reply
New Contributor
Posts: 3
Accepted Solution

Odds ratio table in 9.3

The default behavior for proc logistic in sas 9.3 has changed as described below:

"If the ODDSRATIO statement or CLODDS= option is specified, the default “Odds Ratio” table is no longer

produced, and only the requested results are displayed."

Any idea why this was done? I have code that used to output the oddsratio table using ods, and the CloddsPL table for another variable. Now the oddsratio table no longer exists when I use the CLODDS= option. Is there a way to get this back?


Accepted Solutions
Solution
‎10-21-2011 10:49 AM
SAS Employee
Posts: 4

Odds ratio table in 9.3

Specify CLODDS=BOTH.  The CloddsWald table contains all the results from the default OR table, and you'll still get your CloddsPL results.

CLODDS and ODDSRATIO add several OR tables, and some complaints rolled in about clutter and it being tough to tell which table came from what.  The assumption that anybody specifying CLODDS and/or ODDSRATIO would ignore the default table obviously wasn't quite correct!

View solution in original post


All Replies
Respected Advisor
Posts: 2,655

Odds ratio table in 9.3

It doesn't look like it.  The only workaround I see is to specify each term in a separate ODDSRATIO statement, and get what you need from the ODS table OddsRatiosPL.

Good luck,

Steve Denham

Solution
‎10-21-2011 10:49 AM
SAS Employee
Posts: 4

Odds ratio table in 9.3

Specify CLODDS=BOTH.  The CloddsWald table contains all the results from the default OR table, and you'll still get your CloddsPL results.

CLODDS and ODDSRATIO add several OR tables, and some complaints rolled in about clutter and it being tough to tell which table came from what.  The assumption that anybody specifying CLODDS and/or ODDSRATIO would ignore the default table obviously wasn't quite correct!

New Contributor
Posts: 3

Odds ratio table in 9.3

Thanks for the replies.

Another helpful reply that I got from David Schlotzhauer at SAS is to use the default option for units. This results in all variables being given an estimate.

proc logistic desc data = _temp_;

class cat1 cat2;

model outcome = cont1 cat1 cat2 cont2 /clodds=pl;

units cont1 = 20 / default=1;

run;

☑ This topic is SOLVED.

Need further help from the community? Please ask a new question.

Discussion stats
  • 3 replies
  • 306 views
  • 6 likes
  • 3 in conversation