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: prob>chi2 in probit results


From   Richard Williams <richardwilliams.ndu@gmail.com>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   Re: st: prob>chi2 in probit results
Date   Wed, 30 Jan 2013 21:02:52 -0500

The chi square in the model was 5.87 with 8 d.f. It is therefore
possible that the highly significant effect of one variable was
diluted by the insignificant effects of the 7 others, e.g. If only the
one worthwhile variable was included, it could have a chi square of 5
with one d.f. Which would be highly significant. I wouldn't expect
that to happen but it could.

Sent from my iPad

On Jan 30, 2013, at 8:20 PM, David Hoaglin <dchoaglin@gmail.com> wrote:

> Hi, Nola.
>
> Another way to interpret the result of that likelihood-ratio test is
> that those 8 predictors, working together, do not account for more of
> the variation in pgood than one would expect by chance.
>
> I am not sure why Maarten would consider a model with fewer of those
> predictors.  Such a model would not account for any more of the
> variation in pgood than all 8 of them.  Perhaps that is not what
> Maarten meant.
>
> You have not said enough about the nature of the predictors for me to
> guess whether it would help to express any of them differently.
>
> David Hoaglin
>
> On Tue, Jan 29, 2013 at 2:24 PM, nola l <nola0908@gmail.com> wrote:
>> Hello,
>>
>> Please pardon me if my question looks too silly.
>>
>> I run probit in stata 12 and get following results:
>>
>> . probit pgood age_kid sex gender parent white  totalkid Ed AGE
>>
>> Iteration 0:   log likelihood = -148.89225
>> Iteration 1:   log likelihood = -145.96563
>> Iteration 2:   log likelihood =  -145.9561
>> Iteration 3:   log likelihood =  -145.9561
>>
>> Probit regression                                 Number of obs   =        252
>>                                                           LR chi2(8)
>>    =       5.87
>>                                                           Prob > chi2
>>    =     0.6615
>> Log likelihood =  -145.9561                Pseudo R2       =     0.0197
>>
>> Is it correct that as my results showed Prob > chi2     =     0.6615
>> which is greater than 0.05, then my model is not a good model and I
>> could not really use it?
>>
>> Is it correct that in order to have a acceptable model, I have to get
>> prob>chi2 <0.1? if I got a much bigger number, can I use the model?
>> How should I interpret it?
>>
>> Is there any other command that I can use to test my probit model is
>> usable or not?
>>
>> Thanks in advance for any suggestion that you could give.
>>
>> Best,
>> Nola
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/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/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


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