R/rlargec.R

In evgam: Generalised Additive Extreme Value Models

## r-largest generalised extreme value negative log-likelihood functions

.rlargec.d0 <- function(pars, likdata) {
likdata$y <- as.matrix(likdata$y)
out <- rlargecd0(split(pars, likdata$idpars), likdata$X[[1]], likdata$X[[2]], likdata$X[[3]], likdata$y, likdata$dupid, likdata$duplicate, likdata$args$drop)
out
}

.rlargec.d12 <- function(pars, likdata) {
likdata$y <- as.matrix(likdata$y)
out <- rlargecd12(split(pars, likdata$idpars), likdata$X[[1]], likdata$X[[2]], likdata$X[[3]], likdata$y, likdata$dupid, likdata$duplicate, likdata$args$drop)
out
}

.rlargec.d34 <- function(pars, likdata) {
likdata$y <- as.matrix(likdata$y)
out <- rlargecd34(split(pars, likdata$idpars), likdata$X[[1]], likdata$X[[2]], likdata$X[[3]], likdata$y, likdata$dupid, likdata$duplicate, likdata$args$drop)
out
}

.rlargecfns <- list(d0 = .rlargec.d0, d120 = .rlargec.d12, d340 = .rlargec.d34)

.rlargecfns$q <- .q_gev
.rlargecfns$unlink <- .rlarge_unlink

.rlargecfns$initfn <- function(lst) {
  col <- lst$args$drop + 1
  inits <- c(sqrt(6) * sd(as.matrix(lst$y)[, col]) / pi, .85)
  inits <- c(mean(as.matrix(lst$y)[, col]) - .5772 * inits[1], log(inits[1]), inits[2])
  inits
}