Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: Hausman test
From
Elizabeth Vieira <[email protected]>
To
[email protected]
Subject
Re: st: Hausman test
Date
Fri, 10 Aug 2012 10:55:14 +0100
Thanks for your response.
On Thu, Aug 9, 2012 at 1:25 PM, <[email protected]> wrote:
> An alternative is to calculate mean-deviations and group-specific means for each time-varying covariate, run your -rologit- model with all of them and then use -testparm- to conduct an F test for the additional variables. If F is significant, you retain the fixed effects in your model, which is surely what the Hausman test is there for.
>
> C
>
> -----Original Message-----
> From: Elizabeth Vieira <[email protected]>
> Sender: [email protected]
> Date: Thu, 9 Aug 2012 12:31:18
> To: <[email protected]>
> Reply-To: [email protected]: st: Hausman test
>
> Hello
>
> I am using the rank ordered logistic regression. When I run the
> Hausman test I get the following message"V_b-V_B is not positive
> definite" in some cases and a negative Chi-Square “model fitted on
> these data fails to meet the asymptotic assumptions of the Hausman
> test” in another. The suest test is recommended, but this cannot be
> used with rologit. How I solve this problem?
>
> Best regards
>
> Elizabeth
>
> *
> * 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/
*
* 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/