Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Cragg-Donald Statistic in ivreg2


From   Justin Smith <smithjd5@univmail.cis.mcmaster.ca>
To   statalist@hsphsun2.harvard.edu
Subject   st: Cragg-Donald Statistic in ivreg2
Date   Sun, 11 Jun 2006 18:43:31 -0400

I am trying to calculate some statistics for weak instruments in Stata,
and some of the previous statalist questions seem to indicate that
ivreg2 combined with the "ffirst" option will report the Cragg-Donald
statistic.   

When I type the following (variable names are unimportant):

ivreg2 sperce  rbd11 -  rbd15 (age xrep xstartlate = c2-c5), ffirst

it gives the following output...


First-stage regressions
-----------------------

Summary results for first-stage regressions:

                Shea
Variable      Partial R2      Partial R2       F(  4,196341)    P-value
age             0.0024          0.6450          89191.70         0.0000
xrep            0.0001          0.0090            444.08         0.0000
xstartlate      0.0000          0.0302           1528.72         0.0000



Instrumental variables (2SLS) regression
----------------------------------------

                                                      Number of obs =
  196351
                                                      F(  8,196342) =
  213.66
                                                      Prob > F      =
  0.0000
Total (centered) SS     =  192791.2461                Centered R2   =
 -0.0363
Total (uncentered) SS   =  192793.6585                Uncentered R2 =
 -0.0363
Residual SS             =  199790.6129                Root MSE      =
       1

------------------------------------------------------------------------------
      sperce |      Coef.   Std. Err.      z    P>|z|     [95% Conf.
Interval]
-------------+----------------------------------------------------------------
         age |   .1938617   .1800731     1.08   0.282     -.159075
   .5467984
        xrep |  -1.434245   2.340419    -0.61   0.540    -6.021383
   3.152893
  xstartlate |  -1.292564   2.913459    -0.44   0.657    -7.002838
    4.41771
       rbd11 |  -.0002252   .0007117    -0.32   0.752    -.0016201
   .0011697
       rbd12 |   1.03e-06   2.68e-06     0.38   0.701    -4.23e-06
   6.29e-06
       rbd13 |  -1.70e-09   4.00e-09    -0.43   0.671    -9.53e-09
   6.13e-09
       rbd14 |   1.15e-12   2.57e-12     0.45   0.655    -3.88e-12
   6.18e-12
       rbd15 |  -2.75e-16   5.94e-16    -0.46   0.644    -1.44e-15
   8.90e-16
       _cons |  -1.854898   1.857435    -1.00   0.318    -5.495405
   1.785608
------------------------------------------------------------------------------
Sargan statistic (overidentification test of all instruments):
          1.102
                                                   Chi-sq(1) P-val =
  0.29386
------------------------------------------------------------------------------
Instrumented:  age xrep xstartlate
Instruments:   c2 c3 c4 c5 rbd11 rbd12 rbd13 rbd14 rbd15
------------------------------------------------------------------------------


I expected to see the underidentification diagnostics, but they do not
appear.  Is there and update I need for ivreg2 that will give this
information?

Thank you,

Justin Smith
*
*   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/



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