st: Re: sequential multiple regression

[Kit Baum <baum@bc.edu>]

The subset model will only make sense if the restrictions which it 
imposes on the full model are acceptable to the data; otherwise the 
subset model is misspecified and inconsistent. I don't see that

regress y x1 x2 x3 x4
test x3 x4

is that much work to judge whether the joint test rejects the subset 
model. In fact you could consider the other two submodels (that both 
include x1 x2) with two more lines:

test x3
test x4

so in four lines of code you can test the full model and three 
restricted variants.

Kit

Kit Baum, Boston College Economics
http://ideas.repec.org/e/pba1.html

On Nov 4, 2004, at 2:33 AM, Eugene wrote:


> I am doing a sequential multiple regression where I start with the base
> model: y = a + b1x1 + b2x2
>
> followed by the full model: y = a = b1x1 + b2x2 + b3x3 + b4x4
>
> Is there anyway that I could get STATA to generate the change 
> statistics
> when changing from the base model to the full model? 
*
*   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/