ev.trawl (version 0.1.0)

MarginalGPDLikelihood: Computes Generalised Pareto (log-)likelihood on non-zero exceedances under independence.

Description

Computes Generalised Pareto (log-)likelihood on non-zero exceedances under independence.

Usage

MarginalGPDLikelihood(values, fixed_names, fixed_params, params,
  model_vars_names, logscale = T, transformation = F, n.moments = 4)

Arguments

values

Vector of target values.

fixed_names

Vector of literal names of parameters to keep fixed.

fixed_params

Vector of numerical values of fixed parameters.

params

List of parameters.

model_vars_names

Vector of all parameters names in the model.

logscale

Logical; Default (TRUE) is to use logscale (log-likelihood).

transformation

Boolean to use the Marginal Transform (MT) method.

n.moments

Number of moments the transformed variables should have using the Marginal Transform (MT) method.

Value

Generalised Pareto (log-)likelihood on non-zero exceedances under independence.

Examples

# NOT RUN {
times <- c(1,2,3,4,5)
values <- c(2,0,3,4,0)
delta <- 2
fixed_names <- c("alpha", "kappa")
params <- c(3.4, 0.1)
fixed_params <- c(2.0, 4.3)
model_vars_names <- c("alpha", "beta", "rho", "kappa")
MarginalGPDLikelihood(values, fixed_names, fixed_params, params, model_vars_names)

# }