Statalist The Stata Listserver


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

Re: st: Funky -lrtest- results


From   vwiggins@stata.com (Vince Wiggins, StataCorp)
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Funky -lrtest- results
Date   Wed, 14 Jun 2006 12:34:14 -0500

Danielle Ferry <dferry@nber.org> asks,

> I am comparing 2 models estimated using -arima-. When I run -lrtest-
> afterwards, I get the following:
>
> lrtest v2 v1
> (log likelihoods of null models cannot be compared)
>
> Likelihood-ratio test                                  LR chi2(5)  =      9.11
> (Assumption: v1 nested in v2)                          Prob > chi2 =    0.1048
>
> What does the first message in parentheses mean? Does this mean that
> the results of the LR test shown are invalid?

-lrtest- is trying to be smart and identify null models, to keep you from
potentially making a mistake.  Unfortunately, its rule for identifying null
models is NOT smart.  It looks only at the main regression equation, which in
ARIMA models is often null.

So long as your models are nested, you can ignore this message.

We will look into improving this message or improving when it is shown.


-- Vince
   vwiggins@stata.com

*
*   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