bprobgHsSSBB

It provides the log-likelihood, gradient and observed information matrix for 
penalized or unpenalized maximum likelihood optimization, for the non-random sample selection model case. 
Possible bivariate distributions are 
four Archimedean copulas with two parameters, namely the Clayton-Gumbel, Joe-Gumbel, Joe-Clayton and Joe-Frank. We refer to them
as BB1, BB6, BB7 and BB8, respectively. Their rotated versions are available as well.

Routine for fitting bivariate probit models with semiparametric predictors (including linear and nonlinear effects) in the presence of correlated error equations, endogeneity or sample selection. Bivariate copula models are also supported.

bprobgHsSSBB: Internal Function

Description

Arguments