Re: st: predict

[Chiara Mussida <cmussida@gmail.com>]

Mnay Thanks,
I know that the probability of errorr is too big, but unfortunately
I'm interested on the impact of each variables on the predicted
probability. I must have these outcomes expressed as euqations.

many many thanks
Chiara

On 6 June 2011 12:30, Maarten Buis <maartenlbuis@gmail.com> wrote:
> On Mon, Jun 6, 2011 at 10:55 AM, Chiara Mussida <cmussida@gmail.com> wrote:
>> Likely there is a way to automatically compute predicted prob in STATA
>> (alternative to predict) to get:
>
> You should not do this manually, the chance of making an error is just too big.
>
> However, if you are brave/foolish enough to try, than you can do so,
> and if you do not make an error you will get exactly the same results
> as -predict-:
>
> *------------------------ begin example ------------------------
> sysuse auto, clear
> recode rep78 1/2=3
> mlogit rep78 foreign
>
> predict double pr*, pr
>
> gen double xb1 = _b[4:_cons] + _b[4:foreign]*foreign
> gen double xb2 = _b[5:_cons] + _b[5:foreign]*foreign
>
> gen double denom = 1 + exp(xb1) + exp(xb2)
>
> gen double mp1 = 1/denom
> gen double mp2 = exp(xb1)/denom
> gen double mp3 = exp(xb2)/denom
>
> tab pr1 mp1
> tab pr2 mp2
> tab pr3 mp3
> *-------------------- end example ------------------------------
>
> -- Maarten
>
> --------------------------
> Maarten L. Buis
> Institut fuer Soziologie
> Universitaet Tuebingen
> Wilhelmstrasse 36
> 72074 Tuebingen
> Germany
>
>
> http://www.maartenbuis.nl
> --------------------------
> *
> *   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/
>



-- 
Chiara Mussida
PhD candidate
Doctoral school of Economic Policy
Catholic University, Piacenza (Italy)

*
*   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/