boot.mle: Bootstrap Distribution for Fitted Model

In FAmle: Maximum Likelihood and Bayesian Estimation of Univariate Probability Distributions

Bootstrap Distribution for Fitted Model

Description

This function allows the user to obtain draws from the (parametric) bootstrap distribution of the fitted model's parameters.

Usage

boot.mle(model, B = 200, seed = NULL, start = NULL,
 method = "Nelder-Mead")

Arguments

model	mle object corresponding to the fitted model.
B	Requested number of bootstrap samples.
seed	A seed may be specified (see set.seed)
start	Starting values for the optimization algorithm (if is.null(start)==TRUE, the fitted model's parameters are used as starting values).
method	The optimization method to be used (see optim and mle).

Details

Parametric bootstrap – see References.

Value

model	mle object corresponding to the fitted model.
B	Requested number of bootstrap samples.
seed	The specified seed (see set.seed)
par.star	Array containing realized values from the bootstrap distribution of the maximum likelihood parameter estimators.
gof	The bootstrap distributions of two goodness-of-fit statistics: Anderson-Darling statistic and Pearson's correlation coefficient for the pair ("observed quantiles","fitted quantiles").
p.value	Bootstrap p-values for the two goodness-of-fit statistics.
failure.rate	The proportion of bootstrap samples for which optimization failed using the specified starting values.
total.time	The total amount of time required to generate B bootstrap samples.

References

Davison, A.C., and Hinkley, D.V. (1997). Bootstrap methods and their application. Cambridge University Press.

See Also

mle, Q.conf.int, Q.boot.ci

Examples

data(yarns)
x <- yarns$x
fit.x <- mle(x,'weibull',c(.1,.1))
boot.x <- boot.mle(fit.x,B=10)
boot.x$par.star
boot.x$p.value

FAmle documentation built on March 18, 2022, 5:29 p.m.