Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: -perturb- with more than 14 variable in Probit model
From
Nick Cox <[email protected]>
To
"[email protected]" <[email protected]>
Subject
Re: st: -perturb- with more than 14 variable in Probit model
Date
Tue, 27 Aug 2013 19:10:35 +0100
-perturb- (SSC) was written by John Hendrickx, not evident for some
years among Stata users. Please remember to explain where user-written
programs you refer to come from!
The limit you describe probably arises because of a limit on local
macro length is biting within the program. You could get more
information by using -set trace on-.
Nick
[email protected]
On 27 August 2013 19:03, Afif Naeem <[email protected]> wrote:
> Hello everyone,
>
> I intend to see the impact of multi-colinearity among 13 variables in a Probit model which has a total of 26 covariates in it. I started off with a simple model, with about 10 variables, so I can get the code right for a minimalist model/example. The code was working fine, so I started adding the variable to it to get to my model with 26 covariates. Thats where the issue lies. Stata does not run the code for -perturb- for more than 14 variables. So if I have 15 (or more) variables, it gives me an error saying "unrecognized command: var12" with error code r(199) which does not tell me anything whats wrong with the code. And it is always the 12th independent variable mentioned in the error, regardless of the ordering of the variables. I have the following code for the -perturb- command (I am just writing it for 16 covariates to keep the code simple):
>
> perturb: probit Y_binary var1 var2 var3 var4 var5 var6 var7 var8 var9 var10 var11 var12 var13 var14 var15 var16 [pweight = variable_weight] if in_sample == 1 , poptions ( pvars(var1) prange(0) pfactors( var16 ) pcnttabs(96) )
>
>
> I am wondering if there is some dimensional restrictions that are preventing me to run the code for more than 14 covariates in the Probit model. I got the same problem while running the above mentioned command in both versions of Stata/IC 11 and 12. Any help will be much appreciated.
>
> Thanks,
> Afif
> *
> * For searches and help try:
> * http://www.stata.com/help.cgi?search
> * http://www.stata.com/support/faqs/resources/statalist-faq/
> * http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/