st: RE: ivreg2 & diagnostic statistics

["Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>]

Urska,

It looks like you are using an old -ivreg2- (the newest versions report
the Angrist-Pischke R2 and F stats, which are more useful than Shea's
partial R2).  There have been some computational improvements as well.

If you update your -ivreg2- and -ranktest-, it may solve the problem.

Best wishes,
Mark

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu 
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Urska Kosi
> Sent: 20 May 2011 17:13
> To: statalist@hsphsun2.harvard.edu
> Subject: st: ivreg2 & diagnostic statistics
> 
> Dear statalist,
> 
> I have been encountering some problems with ivreg2 command. 
> In particular with the following diagnostic statistics:
> 
> I have a model with cost of debt as dependent variable. 
> Because loan terms are assumed to be determined jointly, so I 
> use ivreg2 command. I have two endogenous regressors and I 
> use one instrument for the first and 2 instruments for the 
> second. I use cluster option (clustering on firms).
> 
> I would like to get the Cragg-Donald F statistic (using 
> ffirst option). Since I use cluster option with ivreg2, 
> Kleinberg-Paap  F statistic should be reported but I my case 
> even this one cannot be calculated. Is this because I have 2 
> endogenous variables? Any suggestion how to obtain either  
> the Cragg-Donald or Kleinberg-Paap F statistic?
> Similar, I would like to get the Anderson LM statistic (using 
> ffirst option). Since I use cluster option, Kleinberg-Paap LM 
> statistic should be reported but I my case even this one 
> cannot be calculated.
> Is this because I have 2 endogenous variables? Any suggestion 
> how to obtain either  the Anderson or Kleinberg-Paap LM statistic?
> 
> Here is how the output looks like:
> 
> Summary results for first-stage regressions
> -------------------------------------------
> 
> Variable    | Shea Partial R2 |   Partial R2    |  F(  3,  
> 2268)    P-value
> maturity    |     0.0020      |     0.0023      |        3.17 
>       0.0235
> secured     |     0.1845      |     0.2170      |        6.28 
>       0.0003
> 
> NB: first-stage F-stat cluster-robust
> 
> Underidentification tests
> Ho: matrix of reduced form coefficients has rank=K1-1 
> (underidentified)
> Ha: matrix has rank=K1 (identified)
> Kleibergen-Paap rk LM statistic             Chi-sq(2)=.       
>  P-val=     .
> Kleibergen-Paap rk Wald statistic           Chi-sq(2)=.       
>  P-val=     .
> 
> Weak identification test
> Ho: equation is weakly identified
> Kleibergen-Paap Wald rk F statistic                    .
> See main output for Cragg-Donald weak id test critical values
> 
> Weak-instrument-robust inference
> Tests of joint significance of endogenous regressors B1 in 
> main equation
> Ho: B1=0 and overidentifying restrictions are valid
> Anderson-Rubin Wald test     F(3,2268)=13.50     P-val=0.0000
> Anderson-Rubin Wald test     Chi-sq(3)=41.30     P-val=0.0000
> Stock-Wright LM S statistic  Chi-sq(3)=41.80     P-val=0.0000
> 
> NB: Underidentification, weak identification and 
> weak-identification-robust
>     test statistics cluster-robust
> 
> Number of clusters             N_clust  =       2269
> Number of observations               N  =       9027
> Number of regressors                 K  =        172
> Number of instruments                L  =        173
> Number of excluded instruments       L1 =          3
> 
> I appreciate any suggestions or advice about my concerns.
> Thanks a lot,
> Urska
> *
> *   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/
> 


-- 
Heriot-Watt University is a Scottish charity
registered under charity number SC000278.


*
*   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/