Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: problem with perturb


From   KONSTANTARAS KONSTANTINOS <dinokon@otenet.gr>
To   statalist@hsphsun2.harvard.edu
Subject   st: problem with perturb
Date   Sat, 27 Dec 2008 19:29:42 +0200 (EET)

Dear Statalisters,

Seasons Greetings!

I run into a problem in my attempting to run perturb for a multinomial logit application; apparently the problem must have something to do with the way I have defined the covariate to be perturbed, as from what I have found setting trace on the application produces an Invalid Syntax Error as it tries to read the number of pfactors defined as perturbation variables, namely n_par in the ado file; the traced problem is as follows:

....output ommitted..
- forval k=1/`niter' {
= forval k=1/100 {
- forval k=1/`niter' {
- forval i=1/`n_pvar' {
= forval i=1/ {
invalid syntax
 local pv: word `i' of `pvars'
 local pr: word `i' of `prange'
 `how' replace `pv`i'' = `pv'+`perturb'*`pr'
 }

i.e. it does not recognize the number of factors I have input (1), although it has previously produced the initial classification probabilities, adjusted expected table and final reclassification probabilities having recognized that I am using it as factor –see output below.

The command I am issuing is as follows:

perturb2: mlogit depn defone ldef3 sizmc remc recumz  _Iyear_1998 _Iyear_1999 _Iyear_2000 _Iyear_2001 _Iyear_2002 _Iyear_2003 if year>1997?


Any help would be greatly appreciated.

Dino Konstantaras
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index