Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Steven Archambault <archstevej@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: ML Max Non-Linear |
Date | Mon, 31 Jan 2011 20:58:16 -0700 |
This helps, thanks for the clarification. On Mon, Jan 31, 2011 at 1:23 AM, Maarten buis <maartenbuis@yahoo.co.uk> wrote: > --- 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/ > * * 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/