Usage
fit.GPD(data, threshold = NA, nextremes = NA, type = c("ml", "pwm"),
information = c("observed", "expected"), optfunc = c("optim", "nlminb"), ...)
findthreshold(data, ne)
plotTail(object, extend = 2, fineness = 1000, ...)
MEplot(data, omit = 3, labels=TRUE, ...)
xiplot(data, models = 30., start = 15., end = 500., reverse = TRUE,
ci = 0.95, auto.scale = TRUE, labels = TRUE, table = FALSE, ...)
hillPlot(data, option = c("alpha", "xi", "quantile"), start = 15,
end = NA, reverse = FALSE, p = NA, ci = 0.95,
auto.scale = TRUE, labels = TRUE, ...)
plotFittedGPDvsEmpiricalExcesses(data, threshold = NA, nextremes = NA)
showRM(object, alpha, RM = c("VaR", "ES"), extend = 2, ci.p = 0.95,
like.num = 50., ...)
RiskMeasures(out, p)Arguments
alpha
numeric, probability level(s).
auto.scale
logical, whether plot should be automatically
scaled.
ci
numeric, probability for asymptotic confidence
bands.
ci.p
numeric, confidence levels.
data
numeric, data vector or timesSeries.
end
integer, maximum number of exceedances to be
considered.
extend
numeric, extension of plotting range.
fineness
integer, count of points at which to evaluate
the tail estimate.
information
character, whether standard errors should be
calculated with observed or expected
information. This only applies to maximum likelihood type; for
pwm type expected
labels
logical, whether axes shall be labelled.
like.num
integer, count of evaluations of profile likelihood.
type
character, estimation by either ML- or PWM type.
models
integer, count of consecutive gpd models to be
fitted; i.e., the count of different thresholds at which to
re-estimate $\xi$; this many $\xi$ estimates will be
plotted.
ne
integer, count of excesses above the threshold.
nextremes
integer, count of upper extremes to be used.
object
list, returned value from fitting GPD
omit
integer, count of upper plotting points to be
omitted.
optfunc
character, function used for ML-optimization.
option
logical, whether "alpha", "xi" (1 / alpha) or
"quantile" (a quantile estimate) should be plotted.
out
list, returned value from fitting GPD.
p
vector, probability levels for risk measures.
reverse
logical, plot ordered by increasing threshold
or number of extremes.
RM
character, risk measure, either "VaR" or "ES"
start
integer, lowest number of exceedances to be
considered.
table
logical, printing of a result table.
threshold
numeric, threshold value.
...
ellpsis, arguments are passed down to either plot()
or optim() or nlminb().