st: Individual and Joint Variables Orthogonality Test

["Lim Boon Leong" <lboonl@streamyx.com>]

Dear Statalists,

 

I have four included exogenous variables that I wish to carry out the
orthogonality test by using the orthog(varlist_ex) option of xtivreg2.

 

I carry out the test for each of these 4 variables individually and one of
the variables turn out not to be exogenous.  Test results are shown below:

 

-orthog- option:

 

Hansen J statistic (eqn. excluding suspect orthog. conditions):
1.875

                                                   Chi-sq(5) P-val =
0.8661

 

C statistic (exogeneity/orthogonality of suspect instruments):
4.406

                                                   Chi-sq(1) P-val =
0.0358

 

Instruments tested:   ceodual

----------------------------------------------------------------------------
--

Instrumented:         ownexr ownex2r ownex3r L.ldiv L.debt

Included instruments: ceodual ined remcom1 extblock1 L.lmktbk L.roa L.ret

                      L.ltotasset yr04 yr05 yr06

Excluded instruments: lrisk lsale lsale2 ks ks2 ys ik L2.ldiv L.roe
L.ltangible

                      L2.debt

Dropped collinear:    yr02 yr03 yr07

 

The C statistic shows that ceodual cannot be treated as exogenous and
actually it is endogenous.

 

But when I carry out this orthog test for all the four variables jointly, I
obtain the following results:

 

-orthog- option:

 

Hansen J statistic (eqn. excluding suspect orthog. conditions):
0.260

                                                   Chi-sq(2) P-val =
0.8779

 

C statistic (exogeneity/orthogonality of suspect instruments):
6.020

                                                   Chi-sq(4) P-val =
0.1976

 

Instruments tested:   ceodual ined remcom1 extblock1

----------------------------------------------------------------------------
--

Instrumented:         ownexr ownex2r ownex3r L.ldiv L.debt

Included instruments: ceodual ined remcom1 extblock1 L.lmktbk L.roa L.ret

                      L.ltotasset yr04 yr05 yr06

Excluded instruments: lrisk lsale lsale2 ks ks2 ys ik L2.ldiv L.roe
L.ltangible

                      L2.debt

Dropped collinear:    yr02 yr03 yr07

----------------------------------------------------------------------------
--

 

The C statistic shows that all of them are exogenous variables.

 

My question is which test result should I follow, the individual or joint
test results?  And I am having difficulty in finding valid and relevant
instruments for ceodual if it is treated as endogenous variable when
applying two stage least square.

 

Thank you in advance.

 

Best regards,

Boon Leong  

*
*   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/