Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
st: RE: Missing F-statistic using areg, absorb( ) cluster( )
From
"Nick Cox" <[email protected]>
To
<[email protected]>
Subject
st: RE: Missing F-statistic using areg, absorb( ) cluster( )
Date
Wed, 3 Mar 2010 14:38:40 -0000
I'd omit the first two predictors to see what difference that makes.
Nick
[email protected]
Samuel Finkelstein
I have been using the areg command and have recently run into an issue
with missing F-statistics. In particular, I am estimating two very
similar models, where only the dependent variables differ. In
particular, the first model uses the percent change in the amount of a
given product purchased, and the second model uses the percent change
in the amount of a different product purchased. Everything else in
the models stays the same. Given the similarities across the models,
I am having difficulty understanding why one model has an F-statistic
while the other model has a missing F-statistic. I (believe) I
understand this could be a problem of rank, but if that is the case,
shouldn't there be a problem with both models. The output for the two
models is below. If anyone has any advice or any thoughts regarding
this issue, it would be greatly appreciated. The models below cluster
by state (i.e. states in the U.S.) and use yearly fixed effects.
. areg chng_a l.chng_t l2.chng_t popchng l.totf l2.totf l.toti
l2.toti, absorb(year) cluster(state)
Linear regression, absorbing indicators Number of obs =
507
F( 7, 50) =
52.74
Prob > F =
0.0000
R-squared =
0.1724
Adj R-squared =
0.1454
Root MSE =
.38704
(Std. Err. adjusted for 51 clusters in
state)
------------------------------------------------------------------------
------
| Robust
chng_a| Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------
------
chng_t |
L1. | -1.07e-08 9.35e-10 -11.49 0.000 -1.26e-08
-8.87e-09
L2. | -5.11e-09 7.20e-10 -7.10 0.000 -6.56e-09
-3.66e-09
popchng | .8971431 1.431939 0.63 0.534 -1.978991
3.773277
totf |
L1. | -.0011519 .0008915 -1.29 0.202 -.0029426
.0006388
L2. | -.0002194 .0007874 -0.28 0.782 -.001801
.0013622
toti |
L1. | .0000377 .0000311 1.21 0.231 -.0000248
.0001001
L2. | -.0000171 .0000239 -0.72 0.476 -.0000651
.0000308
_cons | .0996357 .0172391 5.78 0.000 .06501
.1342615
-------------+----------------------------------------------------------
------
year | absorbed (10
categories)
. areg chng_b l.chng_t l2.chng_t popchng l.totf l2.totf l.toti
l2.toti, absorb(year) cluster(state)
Linear regression, absorbing indicators Number of obs =
507
F( 5, 50) =
.
Prob > F =
R-squared =
0.4833
Adj R-squared =
0.4664
Root MSE =
.07515
(Std. Err. adjusted for 51 clusters in
state)
------------------------------------------------------------------------
------
| Robust
chng_b| Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------
------
chng_t |
L1. | -1.85e-09 1.68e-10 -10.98 0.000 -2.18e-09
-1.51e-09
L2. | 2.78e-09 1.94e-10 14.36 0.000 2.39e-09
3.17e-09
popchng | .9542675 .2958303 3.23 0.002 .3600749
1.54846
totf |
L1. | .0001632 .0001579 1.03 0.307 -.000154
.0004804
L2. | -.0002011 .0001287 -1.56 0.124 -.0004595
.0000574
toti |
L1. | -.0000115 3.43e-06 -3.34 0.002 -.0000184
-4.58e-06
L2. | 4.40e-06 4.77e-06 0.92 0.361 -5.19e-06
.000014
_cons | .0342539 .0044325 7.73 0.000 .0253508
.0431569
-------------+----------------------------------------------------------
------
year | absorbed (10
categories)
Any thoughts or recommendations would be greatly appreciated.
*
* 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/