R: Getting log generalized determinant of penalty matrices

ldetS {mgcv}

R Documentation

Getting log generalized determinant of penalty matrices

INTERNAL function calculating the log generalized determinant of penalty matrix S stored blockwise in an Sl list (which is the output of Sl.setup).

ldetS(Sl, rho, fixed, np, root = FALSE, repara = TRUE,
      nt = 1,deriv=2,sparse=FALSE)

`Sl`	the output of `Sl.setup`.
`rho`	the log smoothing parameters.
`fixed`	an array indicating whether the smoothing parameters are fixed (or free).
`np`	number of coefficients.
`root`	indicates whether or not to return the matrix square root, `E`, of the total penalty S_tot.
`repara`	if TRUE multi-term blocks will be re-parameterized using `gam.reparam`, and a re-parameterization object supplied in the returned object.
`nt`	number of parallel threads to use.
`deriv`	order of derivative to use
`sparse`	should `E` be sparse?

A list containing:

ldetS: the log-determinant of S.
ldetS1: the gradient of the log-determinant of S.
ldetS2: the Hessian of the log-determinant of S.
Sl: with modified rS terms, if needed and rho added to each block
rp: a re-parameterization list.
rp: E a total penalty square root such that t(E)%*%E = S_tot (if root==TRUE).

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

[Package mgcv version 1.9-1 Index]