st: Re: Ramsey test interpretation

[Kit Baum <baum@bc.edu>]

<>
That is not quite right either, as Jeff Wooldridge clearly explains in  
his undergrad text "Introductory econometrics". What is being tested  
is the functional form of the X's you have included. A significant  
Ramsey test is evidence that, say, including X in levels is  
inadequate, compared to including a nonlinear f(X). As a high-order  
polynomial can approximate any f(X), the test can pick that up.

But the test---despite its poor choice of nomenclature, 'ovtest:  
omitted variables test' in Stata---does NOT test whether you have left  
out variables Z1, Z1^2, etc. where Zs are completely different  
variables.  It does not (and can not) test the maintained hypothesis  
that y = X b + u, with no Zs in X. In other words, the Ramsey test's  
ability to test the specification of a possibly 'short model' is  
limited to considering whether its 'shortness' may be due to leaving  
out powers of Xs, interactions of Xs, etc., and does not really  
consider 'omitted variables' in the broader sense.

Kit

On Sep 19, 2009, at 2:33 AM, Clive wrote:

> Many scholars incorrectly interpret a non-significant F ratio to
> support the conclusion that we have included all the relevant
> variables, when it actually means that there are almost certainly no
> omitted _nonlinear_ variables to include on our model.
>
> In your case, F is significant, so you may well have omitted some
> quadratic, cubic or otherwise nonlinear variables (or, indeed,
> nonlinear transformations of your existing variables) in your model.
> Once you've included them and fit the model, running -ovtest- again
> may produce a different outcome.


Kit Baum   |   Boston College Economics & DIW Berlin   |   http://ideas.repec.org/e/pba1.html
                              An Introduction to Stata Programming   
|   http://www.stata-press.com/books/isp.html
   An Introduction to Modern Econometrics Using Stata  |   http://www.stata-press.com/books/imeus.html

*
*   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/