Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Re: Nonlinear Tobit


From   "Rodrigo A. Alfaro" <ralfaro76@hotmail.com>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: Re: Nonlinear Tobit
Date   Wed, 15 Nov 2006 09:28:56 -0500

Jan,

If your model is y=A/(A+p^r) and p is non-negative
I think that you could solve it using logarithms:

For B=A^(1/r) and x=r*log(p/B) you have
y = 1/(1+(p/B)^r) = 1/(1+exp(r*log(p/B))) = 1/(1+exp(x))

Then for 0<y<1 and z=y/(1-y) you have log(z) = -x

Rodrigo.



----- Original Message ----- 
From: "Jan Heufer" <Steppenwolf_81@gmx.de>
To: <statalist@hsphsun2.harvard.edu>
Sent: Tuesday, November 14, 2006 10:08 PM
Subject: st: Nonlinear Tobit


Hi,

how can I estimate a nonlinear model, when the dependent variable is left- =
and right-censored?

Specifically, I would like to estimate

y =3D A/(A+p^r),

where p is the independent variable, A and r are parameters to be estimated=
, and y is censored at 0 and 1.

I am aware that it is not a problem to estimate the model with NLS, ignorin=
g the censoring. But I'd be really thankful if I could get help on how to u=
se tobit for the model.

Cheers!
--=20
Der GMX SmartSurfer hilft bis zu 70% Ihrer Onlinekosten zu sparen!
Ideal f=FCr Modem und ISDN: http://www.gmx.net/de/go/smartsurfer
*
*   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