gev_optim: GEV optimisation function

In dannyjameswilliams/danielR: Collection of Danny's R Code

GEV optimisation function

Description

Optimise a non-stationary GEV distribution

Usage

gev_optim(
  z,
  muX,
  sigX,
  xiX,
  gr = gev_grad,
  se = TRUE,
  mulink = "identity",
  siglink = "exponential",
  xilink = "identity",
  init = NULL
)

Arguments

z	vector of observed maxima
muX	model matrix for the location parameter
sigX	model matrix for the scale parameter
xiX	model matrix for the shape parameter
gr	gradient function to be used in optim, if NULL, a numerical gradient approximation is used
se	logical; if TRUE, standard error of parameters will be output too
mulink	string describing what link function to use for the location parameter
siglink	string describing what link function to use for the scale parameter
xilink	string describing what link function to use for the shape parameter
init	initial conditions for optimisation

Details

This function uses the optim function with the BFGS method to optimise a parameter vector corresponding to a given set of covariates for each GEV parameter.

Value

vector of gradients for each parameter

dannyjameswilliams/danielR documentation built on Aug. 20, 2023, 3:25 a.m.