Statalist


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

Re: st: How to choose consistent starting values


From   amatoallah ouchen <at.ouchen@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: How to choose consistent starting values
Date   Tue, 1 Sep 2009 15:04:05 +0200

 here is the model

 y*t = xtβ + εt


         yt = y*t - α1  if    y*t> α1

          yt=0        if          α1< y*t< α2

          yt = y*t- α2   if    y*t< α2

 y*t   is the latent variable

with α1< 0  and   α2>0  and ε i.i.d, residuals of the estimation with
variance  σj². The parameters β, α1 α2 and σ are solved by maximising
a  log-likelihood function.
Ln L =   ∑ ln (1/2πσj²)1/2 -∑(yjt+α1j-βjxt)²/2 σj² +
t€ψ1
∑ ln (1/2πσj²)1/2 -∑(yjt+α2j-βjxt)²/2 σj²+
 t€ψ2
∑ln[Ф((α2j-βjxt)/ σj)- Ф((α2j-βjxt)/ σj)]
t€ψ3

As I've told you my principal  problem is on how to find out
consistent  starting points, any help would be highly appreciated !!!

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