msn.dev.grad {sn}R Documentation

Gradient of the deviance for multivariate skew-normal distributions

Description

Computes the gradient of twice the negative profile log-likelihood for multivariate regression models with errors having a multivariate skew-normal distribution.

Usage

msn.dev.grad(param, X, y, freq, trace=F)

Arguments

param a numeric vector of parameter values.
X a matrix of explanatory variables. Missing values (NAs) are not allowed.
y a matrix contaning the response variable. Missing values (NAs) are not allowed.
freq a vector of weights, with length equal to nrow(y).
trace logical value. If trace=T, details are printed. Default value is F.

Details

This is the gradient function of msn.dev, which is the objective function of msn.mle; refer to its documentation for additional details.

Value

a vector contaning the gradient

See Also

msn.mle,msn.dev