st: Re: RE: prchange

["R.E. De Hoyos" <redeho@hotmail.com>]

Andre,

You can use the -prtab- post estimation command to find out the change in 
the conditional probability brought about by a change in one of your 
independent variables. In other words, -prtab- will tell you how much the 
probability of observing an outcome changes with an increase in years of 
experience, covering all possible values of experience.

Rafa

----- Original Message ----- 
From: "Alexander Severinsen" <Alexander.Severinsen@nfh.uit.no>
To: <statalist@hsphsun2.harvard.edu>
Sent: Wednesday, February 15, 2006 1:17 PM
Subject: st: RE: prchange



> Hi Andr�,
>
> Do you mean with age as an independent variable, and not as a dependent? 
> Assuming this you can use the delta command as an option in prchange to 
> specify the amount of discrete change in the independent variable. The 
> default is a 1-unit change.
>
> I believe you can do what asked for this way:
>
> mlogit y age
> prchange age, x(age=20) uncentered delta(20) // from 20 to 40
> prchange age, x(age=40) uncentered delta(20) // from 40 to 60
>
> Options meaning:
> Delta(20)=20 unit change in age
> X(age=20)=specify the starting point of age = 20
> Uncentered = the change should be computed starting at the base value as 
> specified by the x(#) option.
>
> Best wishes,
> Alexander Severinsen
>
>
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu 
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Andr� Paul
> Sent: 15. februar 2006 11:12
> To: statalist@hsphsun2.harvard.edu
> Subject: st: prchange
>
> Dear all,
>
> In the "spost" pakage by Long & Freese, the command "prchange" enables us 
> to measure the effect of discrete changes of the explanatory variables.
>
> I have a question regarding this command:
> Suppose I want to estimate a multinomial logit model, with "age" as a 
> dependant variable, which ranges from 20 to 60. The "prchange" command 
> will give me the effect on the categorial variable when "age" varies from 
> 20 to 60 (If I got that right!).
> Is there a way to obtain this effect for different variations of the 
> variable? i.e if age varies from 20 to 40 and then from 40 to 60? I didn't 
> find the answer in the help file...
>
> Thanks,
> Andr�
>
> _________________________________________________________________
> MSN Messenger : appels gratuits de PC � PC partout dans le monde !
> http://www.imagine-msn.com/Messenger/?locale=fr-fr
>
> *
> *   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/
>
> *
> *   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/
*
*   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/