Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: switching regression


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: switching regression
Date   Mon, 28 Nov 2005 11:47:41 -0000

Eric Cahuzac replied on 24 November to your mail. 

If you are not clear whether your post got through, 
you should check in the Statalist archives. 

Also, see advice at 

http://www.stata.com/support/faqs/res/statalist.html#noanswer

and elsewhere in the Statalist FAQ. 

By the way, comments of the form "I really need your 
help" and "as soon as possible" do not help your case. 
In fact, they are likely to be counter-productive. 

Nick 
n.j.cox@durham.ac.uk 

yjlin3@hotmail.com

> I wrote a letter to you guys last week,
> however I haven't gotten any feedback yet,
> so I wonder that you might not receive my mail.
> Anyway, I think I'd better write a letter to you again.
> I'd like to run switching regressions , but I don't know how.
> I know the command "switchr", but I can't understand the illstration.
> Can you just give me a simple example?
> For example, my dependent variable is y, and my independent 
> variables are b, c, and d.
> Beside, I have 2 variables e,f to be criteria to separate my 
> regression y=
> a0+a1b+a2c+a3d into 4 segments
> first, I'd like to know how I can get the criteria e* and f* 
> which maximize the regression's likelihood.
> then, how can I compare these segments?
> I really need your help, and I hope you can give me any 
> response as soon as possible

*
*   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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index