st: Cragg-Donald minimum eigenvalue statistic

[FredVer <frederiqueverblakt@hotmail.com>]

Dear all,

Again, a queston. Im estimating an ivregress-gmm regression model. Where for
my other regressions use two endogenous regressors iso just one. This second
regressor is the squared term of the first endogenous regressor. Now I've
read some comments and papers, that i need to look at the cragg donald
minimum eigenvalue statistic to see if my instruments are strong. The
f-stats  and the hansen j test are no good diagnostic tests anymore. I
hardly think anyone has the experience here, but the output i now get is
this for diagnostic tests after ivregress gmm:

*estat firststage, all forcenonrobust*

  First-stage regression summary statistics
  --------------------------------------------------------------------------
               |            Adjusted      Partial       Robust
      Variable |   R-sq.       R-sq.        R-sq.       F(3,15)   Prob > F
  -------------+------------------------------------------------------------
       lerner2 |  0.0762      0.0760       0.0193       37.1807    0.0000
      sqlerner |  0.0767      0.0766       0.0100       12.8582    0.0002
  --------------------------------------------------------------------------
  (F statistics adjusted for 16 clusters in councode)


  Shea's partial R-squared
  --------------------------------------------------
               |     Shea's             Shea's
      Variable |  Partial R-sq.   Adj. Partial R-sq.
  -------------+------------------------------------
       lerner2 |     0.0360             0.0359
      sqlerner |     0.0187             0.0185
  --------------------------------------------------


  Minimum eigenvalue statistic = 207.217     

  Critical Values                      # of endogenous regressors:    2
  Ho: Instruments are weak             # of excluded instruments:     3
  ---------------------------------------------------------------------
                                     |    5%     10%     20%     30%
  2SLS relative bias                 |         (not available)
  -----------------------------------+---------------------------------
                                     |   10%     15%     20%     25%
  2SLS Size of nominal 5% Wald test  |  13.43    8.18    6.40    5.45
  LIML Size of nominal 5% Wald test  |   5.44    3.81    3.32    3.09
  ---------------------------------------------------------------------

Another question i have is when i now test for heteroskedasticity (as with
one endogenous regressor there was heteroskedasticity) now with the squared
term included the OLS shows for one sample regression
homoskedasticity....Should I now use some other regression model?

--
View this message in context: http://statalist.1588530.n2.nabble.com/Cragg-Donald-minimum-eigenvalue-statistic-tp7569932.html
Sent from the Statalist mailing list archive at Nabble.com.
*
*   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/