Plots of parameter estimates at various thresholds for peaks over threshold modelling, using the Generalized Pareto or Point Process representation.
tcplot(data, tlim, model = c("gpd","pp"), pscale = FALSE, cmax =
FALSE, r = 1, ulow = -Inf, rlow = 1, nt = 25, which = 1:npar,
conf = 0.95, lty = 1, lwd = 1, type = "b", cilty = 1, vci =
TRUE, xlab, xlim, ylabs, ylims, ask = nb.fig < length(which) &&
dev.interactive(), ...)
A list is invisibly returned. Each component is a matrix with three columns giving parameter estimates and confidence limits.
A numeric vector.
A numeric vector of length two, giving the limits for the thresholds at which the model is fitted.
The model; either "gpd"
(the default) or
"pp"
, for the Generalized Pareto or Point Process
representations respectively.
If TRUE
, then the x-axis gives the threshold
exceedance probability rather than the threshold itself.
Logical; if FALSE
(the default), the models
are fitted using all exceedences over the thresholds. If
TRUE
, the models are fitted using cluster maxima, using
clusters of exceedences derived from clusters
.
Arguments used for the identification of
clusters of exceedences (see clusters
).
Ignored if cmax
is FALSE
(the default).
The number of thresholds at which the model is fitted.
If a subset of the plots is required, specify a
subset of the numbers 1:npar
, where npar
is
the number of parameters, so that npar = 2
when
model = "gpd"
(the default) and npar = 3
when model = "pp"
.
The (pointwise) confidence coefficient for the plotted confidence intervals. Use zero to suppress.
The line type and width of the line connecting the parameter estimates.
The form taken by the line connecting the parameter
estimates and the points denoting these estimates. Possible
values include "b"
(the default) for points joined by
lines, "o"
for overplotted points and lines, and
"l"
for an unbroken line with no points.
The line type of the lines depicting the confidence intervals.
If TRUE
(the default), confidence intervals
are plotted using vertical lines.
Label and limits for the x-axis; if given, these arguments apply to every plot.
A vector of y-axis labels and a matrix of
y-axis limits. If given, ylabs
should have the same
length as which
, and ylims
should have two
columns and length(which)
rows. If the length of
which
is one, then ylims
can be a vector of
length two.
Logical; if TRUE
, the user is asked before
each plot.
Other arguments to be passed to the model fit
function fpot
.
Stuart Coles and Alec Stephenson
For each of the nt
thresholds a peaks over threshold model
is fitted using the function fpot
. When model = "gpd"
(the default), the maximum likelihood estimates for the shape and the
modified scale parameter (modified by subtracting the shape multiplied
by the threshold) are plotted against the thresholds.
When model = "pp"
the maximum likelihood estimates for the
location, scale and shape parameters are plotted against the
thresholds. (The modified scale parameter in the "gpd"
case
is equivalent to the scale parameter in the "pp"
case.)
If the threshold u
is a valid threshold to be used for peaks
over threshold modelling, the parameter estimates depicted should
be approximately constant above u
.
fpot
, mrlplot
,
clusters
tlim <- c(3.6, 4.2)
if (FALSE) tcplot(portpirie, tlim)
if (FALSE) tcplot(portpirie, tlim, nt = 100, lwd = 3, type = "l")
if (FALSE) tcplot(portpirie, tlim, model = "pp")
Run the code above in your browser using DataLab