Learn R Programming

NSM3 (version 1.18)

ferg.df: Ferguson's Estimator

Description

Function to compute an approximation of Ferguson's estimator mu_n.

Usage

ferg.df(x, alpha, mu, npoints,...)

Value

The function returns a vector of length num.points for Ferguson's estimator.

Arguments

x

a vector of data of length n

alpha

the degree of confidence in mu

mu

the prior guess of the unknown P (a pdf)

npoints

the number of estimated points returned

...

all of the arguments needed for mu

Author

Rachel Becvarik

References

See Section 16.2 of Hollander, Wolfe, Chicken - Nonparametric Statistical Methods 3.

Examples

Run this code
##Hollander-Wolfe-Chicken Figure 16.2
framingham<-c(2273, 2710, 141, 4725, 5010, 6224, 4991, 458, 1587, 1435, 2565, 1863)
plot.ecdf(framingham)
lines(sort(framingham),pexp(sort(framingham), 1/2922), lty=3)
temp.x = seq(min(framingham), max(framingham), length.out=100)
lines(temp.x,ferg.df(sort(framingham), 4, npoints=100,pexp,1/2922), col=2, type="s", lty=2)
legend("bottomright",  lty=c(1,3,2), legend=c("ecdf", "prior", "ferguson"), col=c(1,1,2))

Run the code above in your browser using DataLab