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

Re: st: random variable transformation


From   Nick Cox <n.j.cox@durham.ac.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: random variable transformation
Date   Fri, 25 Apr 2003 18:16:02 +0100

Eliana 

> I have to estimate Y which is a variable comprised between 0 and 1.
> 
> I have then transformed into Z=log(Y/(1-Y)) and I have used OLS to
estimate Z.
> 
> The point is: How to get predicted values for Y from predicted
values for Z?

You can back transform algebraically using exp(Z) / (1 + exp(Z)), but
another more general possibility is to use -glm, link(logit)-, which
automatically yields predictions on the scale of Y. You need neither
transformation nor back-transformation. There is still the question of
which error family to use. 

See also 
How does one estimate a model when the dependent variable is a
proportion?
http://www.stata.com/support/faqs/stat/logit.html

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