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

st: RE: Parmeter remapping


From   "Maarten Buis" <M.Buis@fsw.vu.nl>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: Parmeter remapping
Date   Wed, 26 Oct 2005 17:48:07 +0200

A common function for this is the logit function: p is parameter constraint between 0 and 1, the logit is ln{p/(1-p)}

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu]On Behalf Of Ed Blackburne
Sent: woensdag 26 oktober 2005 17:42
To: statalist@hsphsun2.harvard.edu
Subject: st: Parmeter remapping

Dear listers,

I have a ml program where a scalar parameter must be on the unit
interval. Naturally, I have had convergence problems in Stata if I set
`lnf'=. if the constraint is not met.  I know the proper way is to
reparameterize and let Stata search over the real line. (Example
estimate ln(sig) rather than sig in ml normal regression).

Originally I tried the sigmoid function to remap, with no luck. The
routine does work, however, using cloglog to remap.

My question to the numerical optimization experts, is there a preferred
function to transform "all reals" to (0,1) given this context?

Looking forward to any thoughts,

Ed

*
*   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