lax (version 1.1.0)

evd: Loglikelihood adjustment for evd fits

Description

S3 alogLik method to perform loglikelihood adjustment for fitted extreme value model objects returned from the functions fgev and fpot in the evd package. If x is returned from fgev then the call must have used prob = NULL.

Usage

# S3 method for evd
alogLik(x, cluster = NULL, use_vcov = TRUE, ...)

Arguments

x

A fitted model object with certain associated S3 methods. See Details.

cluster

A vector or factor indicating from which cluster the respective loglikelihood contributions from loglik originate. This must have the same length as the vector returned by the logLikVec method for an object like x. If cluster is not supplied (i.e. is NULL) then it is assumed that each observation forms its own cluster. See Details.

use_vcov

A logical scalar. Should we use the vcov S3 method for x (if this exists) to estimate the Hessian of the independence loglikelihood to be passed as the argument H to adjust_loglik? Otherwise, H is estimated inside adjust_loglik using optimHess.

...

Further arguments to be passed to the functions in the sandwich package meat (if cluster = NULL), or meatCL (if cluster is not NULL).

Value

An object inheriting from class "chandwich". See adjust_loglik. class(x) is a vector of length 5. The first 3 components are c("lax", "chandwich", "evd"). The remaining 2 components depend on the model that was fitted. If fgev was used then these components are c("gev", "stat") if nsloc was NULL and c("gev", "nonstat") if nsloc was not NULL. If fpot was used then these components are c("pot", "gpd") if model was "gpd" and c("pot", "pp") if model was "pp".

Details

See alogLik for details.

References

Chandler, R. E. and Bate, S. (2007). Inference for clustered data using the independence loglikelihood. Biometrika, 94(1), 167-183. http://doi.org/10.1093/biomet/asm015

Zeleis (2006) Object-Oriented Computation and Sandwich Estimators. Journal of Statistical Software, 16, 1-16. http://doi.org/10.18637/jss.v016.i09

See Also

alogLik: loglikelihood adjustment for model fits.

Examples

# NOT RUN {
# We need the evd package
got_evd <- requireNamespace("evd", quietly = TRUE)

if (got_evd) {
  library(evd)
  # An example from the evd::fgev documentation
  set.seed(3082019)
  uvdata <- evd::rgev(100, loc = 0.13, scale = 1.1, shape = 0.2)
  M1 <- evd::fgev(uvdata, nsloc = (-49:50)/100)
  adj_fgev <- alogLik(M1)
  summary(adj_fgev)

  # An example from Chandler and Bate (2007)
  owfit <- fgev(ow$temp, nsloc = ow$loc)
  adj_owfit <- alogLik(owfit, cluster = ow$year)
  summary(adj_owfit)

  # An example from the evd::fpot documentation
  set.seed(3082019)
  uvdata <- evd::rgpd(100, loc = 0, scale = 1.1, shape = 0.2)
  M1 <- fpot(uvdata, 1)
  adj_fpot <- alogLik(M1)
  summary(adj_fpot)
  # Fit using the pp model, rather than the gpd
  M1 <- fpot(uvdata, 1, model = "pp", npp = 365)
  adj_fpot <- alogLik(M1)
  summary(adj_fpot)
}
# }