fweibullgpdcon function

MLE Fitting of Weibull Bulk and GPD Tail Extreme Value Mixture Model with Single Continuity Constraint

MLE Fitting of Weibull Bulk and GPD Tail Extreme Value Mixture Model with Single Continuity Constraint

Maximum likelihood estimation for fitting the extreme value mixture model with Weibull for bulk distribution upto the threshold and conditional GPD above threshold with continuity at threshold. With options for profile likelihood estimation for threshold and fixed threshold approach.

fweibullgpdcon(x, phiu = TRUE, useq = NULL, fixedu = FALSE, pvector = NULL, std.err = TRUE, method = "BFGS", control = list(maxit = 10000), finitelik = TRUE, ...) lweibullgpdcon(x, wshape = 1, wscale = 1, u = qweibull(0.9, wshape, wscale), xi = 0, phiu = TRUE, log = TRUE) nlweibullgpdcon(pvector, x, phiu = TRUE, finitelik = FALSE) profluweibullgpdcon(u, pvector, x, phiu = TRUE, method = "BFGS", control = list(maxit = 10000), finitelik = TRUE, ...) nluweibullgpdcon(pvector, u, x, phiu = TRUE, finitelik = FALSE)

Arguments

  • x: vector of sample data
  • phiu: probability of being above threshold (0,1)(0, 1) or logical, see Details in help for fnormgpd
  • useq: vector of thresholds (or scalar) to be considered in profile likelihood or NULL for no profile likelihood
  • fixedu: logical, should threshold be fixed (at either scalar value in useq, or estimated from maximum of profile likelihood evaluated at sequence of thresholds in useq)
  • pvector: vector of initial values of parameters or NULL for default values, see below
  • std.err: logical, should standard errors be calculated
  • method: optimisation method (see optim)
  • control: optimisation control list (see optim)
  • finitelik: logical, should log-likelihood return finite value for invalid parameters
  • ...: optional inputs passed to optim
  • wshape: scalar Weibull shape (positive)
  • wscale: scalar Weibull scale (positive)
  • u: scalar threshold value
  • xi: scalar shape parameter
  • log: logical, if TRUE then log-likelihood rather than likelihood is output

Returns

Log-likelihood is given by lweibullgpdcon and it's wrappers for negative log-likelihood from nlweibullgpdcon

and nluweibullgpdcon. Profile likelihood for single threshold given by profluweibullgpdcon. Fitting function fweibullgpdcon returns a simple list with the following elements

call :optim call
x :data vector x
init :pvector
fixedu :fixed threshold, logical
useq :threshold vector for profile likelihood or scalar for fixed threshold
nllhuseq :profile negative log-likelihood at each threshold in useq
optim :complete optim output
mle :vector of MLE of parameters
cov :variance-covariance matrix of MLE of parameters
se :vector of standard errors of MLE of parameters
rate :phiu to be consistent with evd
nllh :minimum negative log-likelihood
n :total sample size
wshape :MLE of Weibull shape
wscale :MLE of Weibull scale
u :threshold (fixed or MLE)
sigmau :MLE of GPD scale (estimated from other parameters)
xi :MLE of GPD shape
phiu :MLE of tail fraction (bulk model or parameterised approach)
se.phiu :standard error of MLE of tail fraction

Details

The extreme value mixture model with Weibull bulk and GPD tail with continuity at threshold is fitted to the entire dataset using maximum likelihood estimation. The estimated parameters, variance-covariance matrix and their standard errors are automatically output.

See help for fnormgpd for details, type help fnormgpd. Only the different features are outlined below for brevity.

The GPD sigmau parameter is now specified as function of other parameters, see help for dweibullgpdcon for details, type help weibullgpdcon. Therefore, sigmau should not be included in the parameter vector if initial values are provided, making the full parameter vector (wshape, wscale, u, xi) if threshold is also estimated and (wshape, wscale, xi) for profile likelihood or fixed threshold approach.

Negative data are ignored.

Note

When pvector=NULL then the initial values are:

  • MLE of Weibull parameters assuming entire population is Weibull; and
  • threshold 90% quantile (not relevant for profile likelihood for threshold or fixed threshold approaches);
  • MLE of GPD shape parameter above threshold.

Acknowledgments

See Acknowledgments in fnormgpd, type help fnormgpd.

Examples

## Not run: set.seed(1) par(mfrow = c(2, 1)) x = rweibull(1000, shape = 2) xx = seq(-0.1, 4, 0.01) y = dweibull(xx, shape = 2) # Continuity constraint fit = fweibullgpdcon(x) hist(x, breaks = 100, freq = FALSE, xlim = c(-0.1, 4)) lines(xx, y) with(fit, lines(xx, dweibullgpdcon(xx, wshape, wscale, u, xi), col="red")) abline(v = fit$u, col = "red") # No continuity constraint fit2 = fweibullgpd(x, phiu = FALSE) with(fit2, lines(xx, dweibullgpd(xx, wshape, wscale, u, sigmau, xi, phiu), col="blue")) abline(v = fit2$u, col = "blue") legend("topright", c("True Density","No continuity constraint","With continuty constraint"), col=c("black", "blue", "red"), lty = 1) # Profile likelihood for initial value of threshold and fixed threshold approach fitu = fweibullgpdcon(x, useq = seq(0.5, 2, length = 20)) fitfix = fweibullgpdcon(x, useq = seq(0.5, 2, length = 20), fixedu = TRUE) hist(x, breaks = 100, freq = FALSE, xlim = c(-0.1, 4)) lines(xx, y) with(fit, lines(xx, dweibullgpdcon(xx, wshape, wscale, u, xi), col="red")) abline(v = fit$u, col = "red") with(fitu, lines(xx, dweibullgpdcon(xx, wshape, wscale, u, xi), col="purple")) abline(v = fitu$u, col = "purple") with(fitfix, lines(xx, dweibullgpdcon(xx, wshape, wscale, u, xi), col="darkgreen")) abline(v = fitfix$u, col = "darkgreen") legend("topright", c("True Density","Default initial value (90% quantile)", "Prof. lik. for initial value", "Prof. lik. for fixed threshold"), col=c("black", "red", "purple", "darkgreen"), lty = 1) ## End(Not run)

References

http://www.math.canterbury.ac.nz/~c.scarrott/evmix

http://en.wikipedia.org/wiki/Weibull_distribution

http://en.wikipedia.org/wiki/Generalized_Pareto_distribution

Scarrott, C.J. and MacDonald, A. (2012). A review of extreme value threshold estimation and uncertainty quantification. REVSTAT - Statistical Journal 10(1), 33-59. Available from http://www.ine.pt/revstat/pdf/rs120102.pdf

Hu, Y. (2013). Extreme value mixture modelling: An R package and simulation study. MSc (Hons) thesis, University of Canterbury, New Zealand. http://ir.canterbury.ac.nz/simple-search?query=extreme&submit=Go

Behrens, C.N., Lopes, H.F. and Gamerman, D. (2004). Bayesian analysis of extreme events with threshold estimation. Statistical Modelling. 4(3), 227-244.

See Also

dweibull, fgpd and gpd

Other weibullgpd: fitmweibullgpd, fweibullgpd, itmweibullgpd, weibullgpdcon, weibullgpd

Other weibullgpdcon: fweibullgpd, itmweibullgpd, weibullgpdcon, weibullgpd

Other itmweibullgpd: fitmweibullgpd, fweibullgpd, itmweibullgpd, weibullgpdcon, weibullgpd

Other fweibullgpdcon: weibullgpdcon

Author(s)

Yang Hu and Carl Scarrott carl.scarrott@canterbury.ac.nz