Statalist


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

Re: st: z/t-value "."?


From   Markus Köller <[email protected]>
To   [email protected]
Subject   Re: st: z/t-value "."?
Date   Thu, 17 Apr 2008 23:13:14 +0200

I tried some tabs, but couldnŽt find anything weird. The only thing that I know is, that the 2 variables with z-value "." are highly significant. 

To me it seems like the probit doesnt use all observations. It sais number of observations 2900, but in fact, I use 3100. 
And it doesnt give me a "Wald chi2" and no "Prob > chi2". 

Pseudo R2 = 0,4466.

The output sais: 0 failures, 13 successes completely determined.

BUT: It uses around 20 iterations, which seems a little bit much to me and the log pseudolikelihoodvalue of the last 6 iterations is always the same.

Thank you for your time and patience.

Markus.

> -----Ursprüngliche Nachricht-----
> Von: [email protected]
> Gesendet: 17.04.08 22:56:23
> An: [email protected]
> Betreff: Re: st: z/t-value "."?


> --- Markus Köller <[email protected]> wrote:
> > would you shortly explain to me, what you meant by:"Model did not
> > converge?" Does it mean, that my apporach is useless or wrong?
> > As I said, the z-value "." appears only for variables which should
> > have a very significant impact.
> 
> It could mean many things.
> Do you get any comments in the itteration log for the last itteration?
> Do you get any warnings about perfect predictions?
> 
> It is sometimes also helpful to do some tabs of the the dependent
> variable and the variable that is causing problems, just to see if
> something is weird.
> 
> -- Maarten
> 
> 
> -----------------------------------------
> Maarten L. Buis
> Department of Social Research Methodology
> Vrije Universiteit Amsterdam
> Boelelaan 1081
> 1081 HV Amsterdam
> The Netherlands
> 
> visiting address:
> Buitenveldertselaan 3 (Metropolitan), room Z434
> 
> +31 20 5986715
> 
> http://home.fsw.vu.nl/m.buis/
> -----------------------------------------
> 
> 
>       ___________________________________________________________ 
> Yahoo! For Good helps you make a difference  
> 
> http://uk.promotions.yahoo.com/forgood/
> *
> *   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/
> 


_______________________________________________________________________
Jetzt neu! Schützen Sie Ihren PC mit McAfee und WEB.DE. 30 Tage
kostenlos testen. http://www.pc-sicherheit.web.de/startseite/?mc=022220


*
*   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