Methods for objects of class "confint_gev"

Methods for objects of class "confint_gev" returned from confint.evmissing.

Usage

# S3 method for class 'confint_gev'
print(x, ...)

# S3 method for class 'confint_gev'
plot(x, parm = c("mu", "sigma", "xi"), add = TRUE, digits = 2, ...)

Arguments

x

An object inheriting from class "confint_gev", an object returned after a call to confint.evmissing.

...

Further arguments. For print.confint_gev to pass arguments to print. For plot.confint_gev to pass graphical parameters to plot to create the initial plot of the profile log-likelihood.

parm

A character scalar specifying the parameter for which a profile log-likelihood is plotted. Must be a single component of c("mu", "sigma", "xi").

add

A logical scalar. If add = TRUE then the plot is annotated with a horizontal line indicating the critical value for the profile log-likelihood used to calculate the confidence limits, vertical lines indicating the values of these limits and a legend stating the confidence interval.

digits

An integer. Passed to signif to round the confidence limits in the legend, if add = TRUE.

Value

print.confint_gev: the argument x is returned, invisibly.

plot.confint_gev: a numeric vector containing the confidence limits for the parameter requested in parm is returned invisibly.

Details

print.confint_gev. A numeric matrix with 2 columns giving the lower and upper confidence limits for the parameters specified by the argument parm in confint.evmissing. These columns are labelled as (1-level)/2 and 1-(1-level)/2, expressed as a percentage, by default 2.5% and 97.5%.

plot.confint_gev. A plot is produced of the profile log-likelihood for the parameter chosen by parm. Only the parameter values used to profile the log-likelihood in the call to confint.evmissing are included, so if faster = TRUE was used then the plot will not be of a smooth curve but will be triangular in the middle.

Examples

See evmissing_methods.

See also

gev_mle and evmissing_methods.