gt.bpm can be used to test the hypothesis of absence of endogeneity, correlated model equations/errors or non-random sample selection.
Usage
gt.bpm(object)
Arguments
object
A fitted SemiParBIVProbit object as produced by SemiParBIVProbit().
Value
It returns a numeric p-value corresponding to the null hypothesis that $\rho$ is equal to 0.
WARNINGS
This test's implementation is only valid for models with normal errors and probit links.
Details
The gradient test was first proposed by Terrell (2002) and it is based on classic likelihood theory. See Marra et al. (submitted) for full details.
References
Marra G., Radice R. and Filippou P. (submitted), Regression Spline Bivariate Probit Models: A Practical Approach to Testing for Exogeneity.
Terrell G. (2002), The Gradient Statistic. Computing Science and Statistics, 34, 206-215.