To test the hypothesis of whether the means of the
relevant coefficients across the models are equal or
not, following Mahani and Poteshman (2008), we
apply the Seemingly Unrelated Regression (SUR) to
account for the correlation in the error terms of the
two models. We combine the regression estimation
results, the estimated parameters, and the associated
covariance matrix to create one simultaneous robust
estimator of covariance matrix. That is, we stack two
equations to simultaneously estimate the coefficient
and covariance matrix. Wald tests are used to test the
equality of means of the two coefficients concerned.
For the difference of means of the coefficients of
board structure and other indices, the following are
the null and alternative hypotheses: