Transform derivatives wrt mu to derivatives wrt linear...

gamlss.etamu {mgcv}

R Documentation

Transform derivatives wrt mu to derivatives wrt linear predictor

Description

Mainly intended for internal use in specifying location scale models. Let g(mu) = lp, where lp is the linear predictor, and g is the link function. Assume that we have calculated the derivatives of the log-likelihood wrt mu. This function uses the chain rule to calculate the derivatives of the log-likelihood wrt lp. See trind.generator for array packing conventions.

Usage

gamlss.etamu(l1, l2, l3 = NULL, l4 = NULL, ig1, g2, g3 = NULL,
  g4 = NULL, i2, i3 = NULL, i4 = NULL, deriv = 0)

Arguments

`l1`	array of 1st order derivatives of log-likelihood wrt mu.
`l2`	array of 2nd order derivatives of log-likelihood wrt mu.
`l3`	array of 3rd order derivatives of log-likelihood wrt mu.
`l4`	array of 4th order derivatives of log-likelihood wrt mu.
`ig1`	reciprocal of the first derivative of the link function wrt the linear predictor.
`g2`	array containing the 2nd order derivative of the link function wrt the linear predictor.
`g3`	array containing the 3rd order derivative of the link function wrt the linear predictor.
`g4`	array containing the 4th order derivative of the link function wrt the linear predictor.
`i2`	two-dimensional index array, such that `l2[,i2[i,j]]` contains the partial w.r.t. params indexed by i,j with no restriction on the index values (except that they are in 1,...,ncol(l1)).
`i3`	third-dimensional index array, such that `l3[,i3[i,j,k]]` contains the partial w.r.t. params indexed by i,j,k.
`i4`	third-dimensional index array, such that `l4[,i4[i,j,k,l]]` contains the partial w.r.t. params indexed by i,j,k,l.
`deriv`	if `deriv==0` only first and second order derivatives will be calculated. If `deriv==1` the function goes up to 3rd order, and if `deriv==2` it provides also 4th order derivatives.

Value

A list where the arrays l1, l2, l3, l4 contain the derivatives (up to order four) of the log-likelihood wrt the linear predictor.

Author(s)

Simon N. Wood <simon.wood@r-project.org>.