Re: st: "covariates in dataset 2 do not match covariates in dataset1"

[Michelle L Stransky <mlu6@cisunix.unh.edu>]

Yes, I do have a _mj and _mi.  Could these be getting 'lost' between  
regression commands?

Thanks.

Michelle

* Maarten buis <maartenbuis@yahoo.co.uk> [Tue 16 Mar 2010 04:05:22 AM EDT]:

> --- On Mon, 15/3/10, Michelle L Stransky wrote:
> > I'm working with a multiply imputed dataset (so I have 5
> > poverty variables to account for missing values).  I
> > used the -reshape long- command to make each imputed
> > variable it's own case (so now, for each case, I have it
> > listed 5 times each with 1 imputation of income).
> >
> > I'm running a series of regressions, trying to fine tune
> > the predictor variables.  The first regression runs
> > fine but subsequent regressions (whether I use my .do file
> > or copy & paste them into the command window) will not
> > calculate.  I receive an error message stating that
> > "covariates in dataset 2 do not match covariates in
> > dataset1".
> >
> > I checked the file and everything seemed fine; the only
> > things that don't exactly match across imputations of the
> > dataset are the poverty values.  In fact, I've used
> > this exact procedure on 2 other (very similar) datasets with
> > no problems.  It's only been in the last month that I
> > receive this message.
> >
> > What doesn't make sense to me is that the first regression
> > command runs perfectly fine but subsequent commands do not
> > (they're all in exactly the same form-- mim: svy: logistic
> > dv iv1 iv2 iv3).  And if I close Stata and then open it
> > and run the second regression it works.  I've attempted
> > to run this analysis on Stata9 Intercooled and Stata11
> > SE.  It doesn't seem to matter; I receive the same
> > error.
> >
> > Do you have any recommendations or suggestions for me as to
> > why this happens and/or how to fix it?
>
>
> -mim- is looking for a variable _mj to identify the different
> datasets and _mi to identify the observations. Do these exist
> in your dataset?
>
> -- Maarten
>
> --------------------------
> Maarten L. Buis
> Institut fuer Soziologie
> Universitaet Tuebingen
> Wilhelmstrasse 36
> 72074 Tuebingen
> Germany
>
> http://www.maartenbuis.nl
> --------------------------
>
>
>
>
> *
> *   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/


*
*   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/