plot.profile.evd: Plot Profile Deviances and Calculate Profile Confidence
Intervals
Description
Displays profile deviances from a model profiled with
profile.evd, and invisibly returns profile
confidence intervals.
Usage
## S3 method for class 'profile.evd':
plot(x, which = names(x), main = NULL,
ask = nb.fig < length(which) && dev.interactive(), ci = 0.95,
clty = 2, ...)
Arguments
x
An object of class "profile.evd".
which
A character vector giving the parameters for which
the profile deviance is plotted, and for which profile confidence
intervals are calculated. By default all profiled parameters in
x are used.
main
Title of each plot; a character vector, the
same length as which.
ask
Logical; if TRUE, the user is asked before
each plot.
ci
A numeric vector. For each parameter in which
profile confidence intervals are calculated, for each confidence
coefficient in ci (but see Warning).
The intervals are returned invisibly as a list of
vectors/
clty
The line type of the horizontal lines that represent
the profile confidence intervals. To omit the lines set
clty to zero.
...
Other graphics parameters.
Value
Profile devainces are plotted for each parameter in
which.
A list with one element for each parameter in which
is also returned invisibly.
Each element is a vector of length two or a matrix with two
columns, giving the lower and upper limits of each confidence
interval.
Warning
The profile confidence intervals may not have confidence coefficient
ci, because the usual asymptotic properties of maximum
likelihood estimators may not hold.
For the GEV model, the usual asymptotic properties hold when the
shape parameter is greater than $-0.5$ (Smith, 1985).
Fortunately, this is usually the case.
References
Smith, R. L. (1985)
Maximum likelihood estimation in a class of non-regular cases.
Biometrika, 72, 67--90.