BookmarkSubscribeRSS Feed
Demographer
Pyrite | Level 9

I think there is a problem with the class statement using proc MDC (type=mixedlogit). In the results, there is a parameter for the reference value, which is a non-sense. Moreover, this parameter has no standard error. I really don't get why. I know I could just create dummy variables, but I want to be sure that it's not another problem. Thanks.

6 REPLIES 6
Rick_SAS
SAS Super FREQ

Could you share the syntax and output? Perhaps on a SASHELP or a simulated data set so we can all see the problem?

Demographer
Pyrite | Level 9

I work in a secure lab so I can't copy/paste the syntax. The syntax was basic, such as:

Suppose variable "a" has 3 categories.

proc mdc;

class=a;

model y = a b c/

type=mixedlogit nchoice=10 mixed=(normalparm=a);

id id;

run;

The output is normal for the other variables. The only major bug is that there is a parameter for the reference value of the variable "a" and that parameter doesn't have standard error.

Rick_SAS
SAS Super FREQ

Your syntax doesn't correctly specify the reference parameterization.  You need to say

class a(ref='1') / param=ref;

where '1' should be replaced by whatever level you want to use for the reference level.

Demographer
Pyrite | Level 9

Those commands are not supported by PROC MDC. See there:

Rick_SAS
SAS Super FREQ

I apologize. I'd go to tech support with this one.  The documentation doesn't  even mention the syntax you are using.

Demographer
Pyrite | Level 9

I think the documentation is not up to date. The class statement with proc MDC works with SAS 9.4 (with the bug mentionned above), but is not supported with SAS 9.1.

sas-innovate-2024.png

Join us for SAS Innovate April 16-19 at the Aria in Las Vegas. Bring the team and save big with our group pricing for a limited time only.

Pre-conference courses and tutorials are filling up fast and are always a sellout. Register today to reserve your seat.

 

Register now!

What is ANOVA?

ANOVA, or Analysis Of Variance, is used to compare the averages or means of two or more populations to better understand how they differ. Watch this tutorial for more.

Find more tutorials on the SAS Users YouTube channel.

Discussion stats
  • 6 replies
  • 1888 views
  • 3 likes
  • 2 in conversation