Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, is already up and running.

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

Re: st: Urgent-Help

From   Austin Nichols <>
Subject   Re: st: Urgent-Help
Date   Thu, 25 Feb 2010 16:18:00 -0500

A. Sen <> :
Read the help file for -xtoverid- (on SSC) for a generalized Hausman
test that you might adapt to MI.
You can also get the e(b) and e(V) in each imputation for FE and RE
and combine yourself by hand.

On Thu, Feb 25, 2010 at 1:01 PM, A. Sen <> wrote:
> Hi
> I am from the University of Cambridge and this question is regarding STATA11
> and its mi command. I hope you can help me out with this as noone else has
> been able to.
> I have 5 imputed datasets and I am estimating random effects and fixed
> effects models for the panel data I have. To choose between these two, i
> need to perform a hausman specification test. I estimate the RE model first
> and store estimates- estimates store random. I then estimate the FE model
> and go to postestimation and run the hausman test. But the error message is
> - e(b) not found in random. This e(b) and e(v) not found prevents me from
> presenting the results in a table as well. i have tried cmdok as well but
> nothing seems to work. Can you please provide me with a command which allows
> me to carry out this test.
*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index