|
|  |
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
RE: st: FW: Xtivreg and Chi-Square
David,
I think the degrees of freedom for the numerator are not right. I
think the F test is testing the hypothesis that the regressors (not the
fixed effects) are zero, and thus the degrees of freedom should be 4,
isn't that correct?
Thanks,
Sandra
***********************************************
Sandra Mortal
Assistant Professor of Finance
516 Cornell Hall
University of Missouri
Columbia, MO 65211
Office: (573) 884-1684
Fax: (573) 884-6296
***********************************************
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of David M.
Drukker
Sent: Thursday, August 03, 2006 4:49 PM
To: [email protected]
Subject: Re: st: FW: Xtivreg and Chi-Square
Sandra C. Mortal <[email protected]> reported that -xtivreg, fe- is
including the constant in the model-Wald test.
This problem will be fixed in an upcoming ado update.
The example below shows that specifying the -small- option removes the
problem. -test- can can also be used.
-- David
[email protected]
. use abdata
. xtivreg n w k (l.n = l2.n) , fe small
Fixed-effects (within) IV regression Number of obs =
751
Group variable: id Number of groups =
140
R-sq: within = 0.7281 Obs per group: min =
5
between = 0.9305 avg =
5.4
overall = 0.9218 max =
7
F(143,608) =
483.00
corr(u_i, Xb) = 0.3780 Prob > F =
0.0000
------------------------------------------------------------------------
------
n | Coef. Std. Err. t P>|t| [95% Conf.
Interval]
-------------+----------------------------------------------------------
-------------+------
n |
L1. | .2877755 .0585089 4.92 0.000 .1728715
.4026795
w | -.529404 .056022 -9.45 0.000 -.6394241
-.4193838
k | .5075971 .0358554 14.16 0.000 .4371816
.5780126
_cons | 2.605774 .2071001 12.58 0.000 2.199056
3.012493
-------------+----------------------------------------------------------
-------------+------
sigma_u | .38694604
sigma_e | .10764416
rho | .92816972 (fraction of variance due to u_i)
------------------------------------------------------------------------
------
F test that all u_i=0: F(139,608) = 3.12 Prob > F =
0.0000
------------------------------------------------------------------------
------
Instrumented: L.n
Instruments: w k L2.n
. test L.n w k
( 1) L.n = 0
( 2) w = 0
( 3) k = 0
F( 3, 608) = 483.00
Prob > F = 0.0000
*
* 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/
