Hi,

I am doing a regression analysis with cross sectional data and matched pairs design.

In order to check my data for endogeneity, I am using the instrumental variables approach.

The results of these test are set out below. When I do a test for overestimation, STATA returns error 498 (pls refer to the details below). Can anyone pls help me understand the error and how to deal with it?

a) ivregress 2sls FFVsCF RoE RoA MVBV AvgAge l_DirectorsTotalCompensation IndDirPRCT FeDirPRCT ExecDirPRCT CDual MLEV ABig4 NoBM (BoardSize = l_MatchYearTotalAssetsUSDmn)

Instrumental variables (2SLS) regression Number of obs = 685
Wald chi2(13) = 49.51
Prob > chi2 = 0.0000
R-squared = .
Root MSE = .50014

----------------------------------------------------------------------------------------------
FFVsCF | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-----------------------------+----------------------------------------------------------------
BoardSize | .0494414 .0286402 1.73 0.084 -.0066925 .1055752
RoE | .0012472 .00865 0.14 0.885 -.0157064 .0182009
RoA | -.0790826 .09029 -0.88 0.381 -.2560478 .0978825
MVBV | -.0000846 .0011943 -0.07 0.944 -.0024253 .0022561
AvgAge | -.0214312 .0040279 -5.32 0.000 -.0293257 -.0135367
l_DirectorsTotalCompensation | .0145879 .0289758 0.50 0.615 -.0422035 .0713793
IndDirPRCT | -.0311191 .1943675 -0.16 0.873 -.4120724 .3498342
FeDirPRCT | -.3402406 .2005359 -1.70 0.090 -.7332837 .0528025
ExecDirPRCT | .1457449 .3872501 0.38 0.707 -.6132515 .9047412
CDual | .0811385 .0395101 2.05 0.040 .0037001 .1585769
MLEV | .0125969 .0867694 0.15 0.885 -.157468 .1826618
ABig4 | -.1228282 .0582887 -2.11 0.035 -.237072 -.0085845
NoBM | .0042921 .0048245 0.89 0.374 -.0051637 .0137479
_cons | 1.182019 .35235 3.35 0.001 .4914255 1.872612
----------------------------------------------------------------------------------------------
Instrumented: BoardSize
Instruments: RoE RoA MVBV AvgAge l_DirectorsTotalCompensation IndDirPRCT FeDirPRCT
ExecDirPRCT CDual MLEV ABig4 NoBM l_MatchYearTotalAssetsUSDmn

b) estat endogenous

Tests of endogeneity
Ho: variables are exogenous

Durbin (score) chi2(1) = 7.27856 (p = 0.0070)
Wu-Hausman F(1,670) = 7.19564 (p = 0.0075)

c) estat firststage

First-stage regression summary statistics
--------------------------------------------------------------------------
| Adjusted Partial
Variable | R-sq. R-sq. R-sq. F(1,671) Prob > F
-------------+------------------------------------------------------------
BoardSize | 0.4724 0.4622 0.1160 88.0617 0.0000
--------------------------------------------------------------------------


Minimum eigenvalue statistic = 88.0617

Critical Values # of endogenous regressors: 1
Ho: Instruments are weak # of excluded instruments: 1
---------------------------------------------------------------------
| 5% 10% 20% 30%
2SLS relative bias | (not available)
-----------------------------------+---------------------------------
| 10% 15% 20% 25%
2SLS Size of nominal 5% Wald test | 16.38 8.96 6.66 5.53
LIML Size of nominal 5% Wald test | 16.38 8.96 6.66 5.53
---------------------------------------------------------------------


d) estat overid
no overidentifying restrictions
r(498);