Statalist The Stata Listserver

[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: conformability error with micombine logit

From   "Nick Cox" <>
To   <>
Subject   st: RE: conformability error with micombine logit
Date   Wed, 9 May 2007 13:16:22 +0100

My only suggestion is that you follow advice and show us
exactly what you typed, plainly and directly. 

Also, comments that this used to work, but doesn't any 
longer, are based on a misconception, or at any rate are
not informative. Software is not like some 
machine that may start malfunctioning because some component
has broken or is becoming defective. Unless you changed
your set-up by installing extra stuff (in which case
you should tell us exactly what has changed) the programs
you are using are presumably just the same as before. 

Finally, I note a trade-off. You are saving yourself some
time by using lower case throughout. This poses some extra 
strain on those who read your posts. 


> in the past, i've been able to run this exact logistic fine, 
> but for the past two days i've been getting "conformability 
> error r(503)".
> i can run micombine logits fine with other dependent vars and 
> the same independent vars.  also, i figured out that when i 
> take one of the independent variables (a dummy for native 
> american), the regression runs fine.  or if i run it without 
> the micombine prefix, it runs fine.
> i tried using the mim prefix and get the following error:
> covariates in analysis of imputed dataset 3 do not match 
> those of imputed dataset 1
> r(498);
> i haven't been able to get mim to work with any of my binary 
> logits and have just been using it for my mlogits.
> any suggestions on why i'm getting the comformability error?  
> has anyone else received this when trying to run an mcombine logit?

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index