Re: st: ML Max Non-Linear

[Maarten buis <maartenbuis@yahoo.co.uk>]

--- On Mon, 31/1/11, Steven Archambault wrote:
> I have a clarification question on some non-linear
> programming, which includes defining my log-likelihood
> function as follows:
> 
> -quietly gen double `sigma'=exp(`eta')-
> -quietly replace
> `lnL'=-.5*ln(2*_pi)-.5*ln((`sigma')^2)-.5*(yindx-`xb')^2/(`sigma')^2-
> 
> I am ensuring sigma is positive.
> 
> The results show eta as not significant (see below). I
> assume this is not a big deal, just says that the null
> that eta is not equal to 0. Here is the question In 
> this case, do I just assume sigma is 1?

In your model eta=ln(sigma), so your null hypothesis is:

ln(sigma) = 0, so sigma = exp(0) = 1

This is an hypothesis, not an assumption.

Hope this helps,
Maarten

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

http://www.maartenbuis.nl
--------------------------


      

*
*   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/