Bookmark and Share

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: Wald test in Random Coefficient Model


From   Shikha Sinha <shikha.sinha414@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Wald test in Random Coefficient Model
Date   Thu, 24 Feb 2011 16:44:37 -0500

Thank you very much for your response. It is WHO "Health Behaviour in
school-aged children " data.

May I also ask you the interpretation of "LR test vs. logistic
regression:     chi2(3) =  2132.55   Prob > chi2 = 0.0000". in the
following output.

------------------------------------------------------------------------------
  Random-effects Parameters  |   Estimate   Std. Err.     [95% Conf. Interval]
-----------------------------+------------------------------------------------
wp5: Unstructured            |
                   var(drel) |    .037479   .0209765      .0125136    .1122521
                  var(_cons) |   .1952712   .0515963      .1163391    .3277558
             cov(drel,_cons) |  -.0052752   .0251633     -.0545943    .0440439
------------------------------------------------------------------------------
LR test vs. logistic regression:     chi2(3) =  2132.55   Prob > chi2 = 0.0000

Thanks,
Shikha

On Thu, Feb 24, 2011 at 4:19 PM, Joerg Luedicke
<joerg.luedicke@gmail.com> wrote:
>> My question is How do test the significance of randome coefficient
>> variance at country level? I saw a few papers using wald statistics.
>> Could you suggest how to get the walt statistic from the above output?
>>  Is it this - Wald chi2(11)      =   2324.18?
>
> I don't know what kind of Wald "test" those others are using. If that
> is not clear from the paper you may bug the authors about that.
> Anyway, the one that you mentioned (Wald chi2(11)      =   2324.18) is
> no such "test" (this one is just what is sometimes referred to as an
> "omnibus" test, basically telling you that at least one coefficient
> from your model differs significantly from zero...).  However, what
> you could do to check if the variation of the effect across country is
> sort of considerable is to do a likelihood ratio test in which you
> compare the model fit with and without the random coefficient (-help
> lrtest-). If the model fits better in the case where variation of the
> coefficient is allowed across countries it might make sense to further
> investigate that. One thing that is definitely helpful in this context
> is a caterpillar plot in which you plot the effect for each country
> including confidence bounds (and e.g. order it from smallest to
> biggest). From that you can immediately see to what extent the
> coefficient varies across countries.
>
> May I ask you what data that is? Is this some ISSP stuff?
>
> HTH,
>
> J.
>
> *
> *   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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index