Re: st: RE: Estimation binary-continuous outcomes model

[Bond Tiger <bond0910@ymail.com>]

Hi Nick,

Thank you for the references that you provided.

I have another question,

(1) Do you know, if there is any procedure in STATA to estimate a mixed Tobit 
and Probit model in simultaneous setup? 

Now, about the issues you asked for clarification:

About my 1st Quo in the earlier post, I was just curious, whether CDSIMEQ is 
used quite often for estimation in this type of models or more efficient 
estimation methods are in use. So I put up the question. 

Also because, I am doing an econometric estimation, so inadvertently I phrased 
it as an econometric issue. But yes, certainly my question is related to 
statistics.

Anyways, thank you very much for replying to my query. 

Best Wishes,

Bond 




----- Original Message ----
From: Nick Cox <n.j.cox@durham.ac.uk>
To: "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Sent: Thu, September 23, 2010 3:04:48 PM
Subject: st: RE: Estimation binary-continuous outcomes model

Bond Tiger: 

-cdsimeq- is by O.M.G. Keshk: 

SJ-4-4  st0038_1  . . . . . . . . . . . . . . . .  Software update for cdsimeq
        (help cdsimeq if installed) . . . . . . . . . . . . . . O. M. G. Keshk
        Q4/04   SJ 4(4):491
        updated to allow all postestimation commands except
        lrtest and suest

SJ-3-2  st0038  CDSIMEQ: A program to implement two-stage probit least squares
        (help cdsimeq if installed) . . . . . . . . . . . . . . O. M. G. Keshk
        Q2/03   SJ 3(2):157--167
        two-stage probit least squares estimation for simultaneous
        equation models in which one of the endogenous variables
        is continuous and the other is dichotomous

. findit cdsimeq

yields only the references above. 

The author is not to my knowledge a member of Statalist or least did not respond 
to your first question on this subject on 21 September. 


Q.1 might benefit from rephrasing in terms of the problem you want to solve, 
rather than the method you think may be out-of-date, or out of fashion. I don't 
know quite what makes your question econometric rather than statistical, so some 
clarity on that might widen the audience, or make it more precise, as the case 
may be. 


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

P.S. "No, Groucho is not my real name. I am breaking it in for a friend."

Bond Tiger 23 Sept 2010
=======================

Please anybody tell me about the stata code CDSIMEQ.

(1) Is this code, quite prevalent in use in current researches or since it is 
based on old econometric technique proposed by Maddala (1983), is now not so 
much in use and rather other advanced methods are in use? If so, what are the 
methods in use?

(2) How to calculate the marginal effects from the probit model using this 
procedure?

Bond Tiger 21 Sept 2010
======================

Can anyone help answering the following questions:

(1) Is there any module for calculating MARGINAL EFFECTS after CDSIMEQ?
(2) How to test for endogeneity of a regressor with CDSIMEQ?
(3) Is there any module like CDSIMEQ for Tobit models (with a continuous and a 
binary endogenous variables)? 


*
*   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/