| |
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: RE: Pooled OLS
I was a bit wrong in my last email.
When you pool a data set, you want your coefficient estimates to be
BLUE.
Assume there is a possible "break" in the data. There are effectively
two equations.
Y1 = a1+b1*X1+c1*X2+U1
Y2 = a2+b2*X1+c2*X2+U2
Effectively, you want to determine if the following are true:
a1 = a2
b1 = b2
c1 = c2
If the following tests are true, then you can pool the data set. If the
tests are false, there may be a way to create a dummy variable(s) to
control for the "break" in the data so you can still pool the data.
These Wald tests allow you to determine if the estimated coefficients
are unbiased (BLU).
You may also want to test for constant variance if you pool the data (U1
= U2). If the data set does not have constant variance, then you need
to correct for the inconsistent variance so you can conduct significance
tests. This allows you to determine if the estimated coefficients are
unbiased and efficient (BLUE).
The following site is very useful:
http://www.stata.com/support/faqs/stat/chow3.html
Sorry for the confusion.
Hope this helps.
Justin White
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Joanne
Marshall
Sent: Wednesday, January 17, 2007 7:20 AM
To: [email protected]
Subject: st: Pooled OLS
Dear fellow,
After reading excess amount of literature on pooled OLS regression, I
still
do not quite understand why its OLS might not be BLUE.
Can someone please interpret?
Thank you.
Cheers
Jo
_________________________________________________________________
MSN Hotmail is evolving - check out the new Windows Live Mail
http://ideas.live.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/
*
* 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/