Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: RE: Missing F-statistic using areg, absorb( ) cluster( )


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
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 
n.j.cox@durham.ac.uk 

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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index