Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: ivprob problem


From   "Nick Cox" <[email protected]>
To   <[email protected]>
Subject   st: RE: ivprob problem
Date   Wed, 20 Nov 2002 22:54:01 -0000

Harald Seider
> 
> when running Joe Harness's ivprob procedure I always get an 
> error message 
> (conformability error). I have no clue why this procedure 
> won't work.
> I have about 250.000 observations which might be an issue. 
> However, the 
> error will also show up if we reduce the number of 
> observations down to a 
> few hundred.
> The next issue might be the fact that many of our variables 
> are dummies 
> (d*, transfer, diag* , dum*).
> Has anyone out there faced a similar problem before and 
> know a solution for it?
> 
> ivprob mort30 ,
> 	endog(sqrtvol)
> 	exog(procptca d0* d1* d2* d3* d5*  transfer
> 	          diagalkalos_x diagamisequ diagarterem 
> diagasthma_x diagatheros
>           		diagbbblk_x diagcardiomeg_x diagchf 
> diagchrglom diagchrlive_x 
> diagchrpulh_x
>           		diagchrrena diagcnsdis diagcollvas 
> diagcopd diagdbtcmp 
> diagdbtuncm diagdeg2avb
>           		diagdement  diagepilep diagfractur 
> diaggihemor diaghismali_x 
> diaghrsecma
>           		diaghtb diaghthrtf_x diaghyperli_x 
> diagincorsy_x diaglrpmali_x 
> diagmitvalv_x
>           		diagnutri_x diagostarth_x diagothcva2 
> diagothvalv_x diagprcabg 
> diagprevmi_x
>           		diagpsychos diagshock_x diagsknulcr 
> diagsyncop_x diagthyroid 
> diagunstangina
>                    	dum*)
> 	iv (dum*  pat3rad0to3   pat4rad3to7  pat7rad7to14  
> pat4rad14to18 
> pat4rad18to22 hosprad*) ;
> 

Here is my wild guess (I don't profess to understand
the econometrics here): 

Within -ivprob- one or more of your covariates
is being dropped because of collinearity when 
it runs 

	probit `depvar' `fitted' `exog' if `touse'

When the matrix of means is being decorated
with column names, there are fewer columns
than names, and Stata squawks. 

There is a possibility that there is 
something for Joe to fix using -_rmcoll-. 

Nick 
[email protected] 

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index