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

Re: st: probit


From   David Greenberg <dg4@nyu.edu>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: probit
Date   Thu, 17 Jun 2004 19:25:10 -0400

You should not use probit (or logit). Treat your dependent variable as continuous, ranging between 0 and 1. To linearize the equation, create a new variable, ln[PIT/(1-PIT)] and treat that as a dependent variable. This will work as long as you don't have values in which PIT is 0 or 1. If you do, you may need to use a different transformation. 
- David Greenberg, Sociology Department, New York University

----- Original Message -----
From: Hadiye Aslan <ha38@cornell.edu>
Date: Thursday, June 17, 2004 4:06 pm
Subject: st: probit

> I have a dependednt variable "probability of informed trading" 
> (PIT) and I
> want to find the determinants of this variable. Since PIT is a 
> probability i
> thought i need to run Probit regressions. The problem is that  I 
> have around
> 40,000 firm-years and PIT can take any value betwen 0 and 1 in 
> each of these
> firm-years. The probit examples that i saw usually have more than 
> 2 and less
> than 10 categories. But, in my case i would say i have around 40,000
> categories.. how do i deal with this problem?
> 
> Once i do the probit analysis in a right way how could I find the 
> marginaleffects?
> I would be more than grateful, if soemone could help me.
> thanks,
> -h.
> <')" >statalist@hsphsun2.harvard.edu>
> 
> *
> *   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/



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