Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: st: Fwd: exponential regression in stata


From   Nick Cox <n.j.cox@durham.ac.uk>
To   "'statalist@hsphsun2.harvard.edu'" <statalist@hsphsun2.harvard.edu>
Subject   RE: st: Fwd: exponential regression in stata
Date   Wed, 19 Jan 2011 15:46:26 +0000

"ln y vs x" and "ln on both sides" are exactly the same proposal as ln(exp(x)) is nothing but x. 

I meant most assumptions about functional form. 

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

Bastian Steingros (a.k.a. Markus) 

Dear Nick,
thanks for the comment.
what kind of details would be helpful in your eyes? About the var's? Their distribution?
to the classical method: you recommend ln(y) on x. Justina recommend ln on both sides. To obtain the e function, I should use your approach right?

> Von: Nick Cox <njcoxstata@gmail.com>

> There are several ways to do this and differing opinions on their
> relative merits. The classical method is a regression of ln y on x.
> But I'd try -glm, link(log)- instead. There might be reasons to use
> -nl- instead, but -nl- can be awkward.
> 
> There are lots of ways to extend the model you mention to three
> predictors, but without any details helpful comment is difficult.
 
> On Wed, Jan 19, 2011 at 9:38 AM, Bastian Steingros <Steingros@gmx.de>
> wrote:
> 
> > I want to run a exp. regression in stata. But I have not found any
> instructions in the help options. do you have any ideas what to enter in stata
> to obtain such a regression model?
> >
> > I mean by exp. function something like : y=0,343 * e^(22,33*x)
> >
> > [my data set has 1 dep. var. and 3 indep. var.'s --> is such a model
> also possible for 3 x-var's?]

*
*   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/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index