eqgevd.Rd
Estimate quantiles of a generalized extreme value distribution.
eqgevd(x, p = 0.5, method = "mle", pwme.method = "unbiased",
tsoe.method = "med", plot.pos.cons = c(a = 0.35, b = 0), digits = 0)
a numeric vector of observations, or an object resulting from a call to an
estimating function that assumes a generalized extreme value distribution
(e.g., egevd
). If x
is a numeric vector,
missing (NA
), undefined (NaN
), and infinite (Inf
, -Inf
)
values are allowed but will be removed.
numeric vector of probabilities for which quantiles will be estimated.
All values of p
must be between 0 and 1. The default value is p=0.5
.
character string specifying the method to use to estimate the location, scale, and
threshold parameters. Possible values are
"mle"
(maximum likelihood; the default),
"pwme"
(probability-weighted moments), and
"tsoe"
(two-stage order-statistics estimator of Castillo and Hadi (1994)).
See the DETAILS section of the help file for egevd
for more
information on these estimation methods.
character string specifying what method to use to compute the
probability-weighted moments when method="pwme"
. The possible values are
"ubiased"
(method based on the U-statistic; the default), or
"plotting.position"
(method based on the plotting position formula).
See the DETAILS section of the help file for egevd
for more
information. This argument is ignored if method
is not equal to
"pwme"
.
character string specifying the robust function to apply in the second stage of
the two-stage order-statistics estimator when method="tsoe"
. Possible
values are "med"
(median; the default), and "lms"
(least median of squares). See the DETAILS section of the help file for
egevd
for more information on these estimation methods.
This argument is ignored if method
is not equal to "tsoe"
.
numeric vector of length 2 specifying the constants used in the formula for the
plotting positions when method="pwme"
and pwme.method="plotting.position"
. The default value is plot.pos.cons=c(a=0.35, b=0)
. If this vector has a names attribute with
the value c("a","b")
or c("b","a")
, then the elements will be
matched by name in the formula for computing the plotting positions. Otherwise,
the first element is mapped to the name "a"
and the second element to the
name "b"
. See the DETAILS section of the help file for egevd
for more information. This argument is used only if method="tsoe"
, or if
both method="pwme"
and pwme.method="plotting.position"
.
an integer indicating the number of decimal places to round to when printing out
the value of 100*p
. The default value is digits=0
.
The function eqgevd
returns estimated quantiles as well as
estimates of the location, scale and threshold parameters.
Quantiles are estimated by 1) estimating the location, scale, and threshold
parameters by calling egevd
, and then 2) calling the function
qgevd
and using the estimated values for
location, scale, and threshold.
If x
is a numeric vector, eqevd
returns a
list of class "estimate"
containing the estimated quantile(s) and other
information. See estimate.object
for details.
If x
is the result of calling an estimation function, eqevd
returns a list whose class is the same as x
. The list
contains the same components as x
, as well as components called
quantiles
and quantile.method
.
Castillo, E., and A. Hadi. (1994). Parameter and Quantile Estimation for the Generalized Extreme-Value Distribution. Environmetrics 5, 417–432.
Forbes, C., M. Evans, N. Hastings, and B. Peacock. (2011). Statistical Distributions. Fourth Edition. John Wiley and Sons, Hoboken, NJ.
Greenwood, J.A., J.M. Landwehr, N.C. Matalas, and J.R. Wallis. (1979). Probability Weighted Moments: Definition and Relation to Parameters of Several Distributions Expressible in Inverse Form. Water Resources Research 15(5), 1049–1054.
Hosking, J.R.M. (1984). Testing Whether the Shape Parameter is Zero in the Generalized Extreme-Value Distribution. Biometrika 71(2), 367–374.
Hosking, J.R.M. (1985). Algorithm AS 215: Maximum-Likelihood Estimation of the Parameters of the Generalized Extreme-Value Distribution. Applied Statistics 34(3), 301–310.
Hosking, J.R.M., J.R. Wallis, and E.F. Wood. (1985). Estimation of the Generalized Extreme-Value Distribution by the Method of Probability-Weighted Moments. Technometrics 27(3), 251–261.
Jenkinson, A.F. (1969). Statistics of Extremes. Technical Note 98, World Meteorological Office, Geneva.
Johnson, N. L., S. Kotz, and N. Balakrishnan. (1995). Continuous Univariate Distributions, Volume 2. Second Edition. John Wiley and Sons, New York.
Landwehr, J.M., N.C. Matalas, and J.R. Wallis. (1979). Probability Weighted Moments Compared With Some Traditional Techniques in Estimating Gumbel Parameters and Quantiles. Water Resources Research 15(5), 1055–1064.
Macleod, A.J. (1989). Remark AS R76: A Remark on Algorithm AS 215: Maximum Likelihood Estimation of the Parameters of the Generalized Extreme-Value Distribution. Applied Statistics 38(1), 198–199.
Prescott, P., and A.T. Walden. (1980). Maximum Likelihood Estimation of the Parameters of the Generalized Extreme-Value Distribution. Biometrika 67(3), 723–724.
Prescott, P., and A.T. Walden. (1983). Maximum Likelihood Estimation of the Three-Parameter Generalized Extreme-Value Distribution from Censored Samples. Journal of Statistical Computing and Simulation 16, 241–250.
Two-parameter extreme value distributions (EVD) have been applied extensively since the 1930's to several fields of study, including the distributions of hydrological and meteorological variables, human lifetimes, and strength of materials. The three-parameter generalized extreme value distribution (GEVD) was introduced by Jenkinson (1955) to model annual maximum and minimum values of meteorological events. Since then, it has been used extensively in the hydological and meteorological fields.
The three families of EVDs are all special kinds of GEVDs. When the shape
parameter \(\kappa=0\), the GEVD reduces to the Type I extreme value (Gumbel)
distribution. (The function zTestGevdShape
allows you to test
the null hypothesis \(H_0: \kappa=0\).) When \(\kappa > 0\), the GEVD is
the same as the Type II extreme value distribution, and when \(\kappa < 0\)
it is the same as the Type III extreme value distribution.
Hosking et al. (1985) compare the asymptotic and small-sample statistical properties of the PWME with the MLE and Jenkinson's (1969) method of sextiles. Castillo and Hadi (1994) compare the small-sample statistical properties of the MLE, PWME, and TSOE. Hosking and Wallis (1995) compare the small-sample properties of unbaised \(L\)-moment estimators vs. plotting-position \(L\)-moment estimators. (PWMEs can be written as linear combinations of \(L\)-moments and thus have equivalent statistical properties.) Hosking and Wallis (1995) conclude that unbiased estimators should be used for almost all applications.
# Generate 20 observations from a generalized extreme value distribution
# with parameters location=2, scale=1, and shape=0.2, then compute the
# MLEs of location, shape,and threshold, and estimate the 90th percentile.
# (Note: the call to set.seed simply allows you to reproduce this example.)
set.seed(498)
dat <- rgevd(20, location = 2, scale = 1, shape = 0.2)
eqgevd(dat, p = 0.9)
#>
#> Results of Distribution Parameter Estimation
#> --------------------------------------------
#>
#> Assumed Distribution: Generalized Extreme Value
#>
#> Estimated Parameter(s): location = 1.6144630
#> scale = 0.9867007
#> shape = 0.2632493
#>
#> Estimation Method: mle
#>
#> Estimated Quantile(s): 90'th %ile = 3.289912
#>
#> Quantile Estimation Method: Quantile(s) Based on
#> mle Estimators
#>
#> Data: dat
#>
#> Sample Size: 20
#>
#Results of Distribution Parameter Estimation
#--------------------------------------------
#
#Assumed Distribution: Generalized Extreme Value
#
#Estimated Parameter(s): location = 1.6144631
# scale = 0.9867007
# shape = 0.2632493
#
#Estimation Method: mle
#
#Estimated Quantile(s): 90'th %ile = 3.289912
#
#Quantile Estimation Method: Quantile(s) Based on
# mle Estimators
#
#Data: dat
#
#Sample Size: 20
#----------
# Clean up
#---------
rm(dat)