[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: weighted least squares with time dummy variables
|
From |
=?ks_c_5601-1987?B?seggtbXH0Q==?= <[email protected]> |
|
To |
[email protected] |
|
Subject |
st: weighted least squares with time dummy variables |
|
Date |
Tue, 23 Jan 2007 17:12:46 +0900 |
Dear statalistserve,
I'm trying to run weighted least square with time dummy variables.
I have 6 time periods (py1-py6), and created 5 time dummies (pydu2-pydu6:
pydu1 can be a reference group).
So, basic regression model is:
reg fer awer unemrt trearn pydu2-pydu6
For the weighted least square, I multiplied weight (=w1) by all DVs and IVs
(including time dummies).
Weigthed model is:
reg w1fer w1awer w1unemrt w1trearn w1pydu2-w1pydu6
My question is that, when I change the reference group, the regression
coefficients also change.
Could you please explain why?
