rASS: Adaptive Slice Sampling Algorithm With Stepping-Out...

In teacher7738/AdapSamp: Adaptive Sampling Algorithms

Description

Adaptive Slice Sampling Algorithm With Stepping-Out Procedures rASS generates a sequence by Adaptive Slice Sampling Algorithm With Stepping-Out Procedures.

Usage

rASS(n, x0 = 0, formula, w = 3)

Arguments

n	Desired sample size
x0	Initial value
formula	Target density function p(x)
w	Length of the coverage interval

References

Neal R M. Slice sampling - Rejoinder[J]. Annals of Statistics, 2003, 31(3):758-767.

Examples

#Expotential distribution
x<-rASS(500,-1,"1.114283*exp(-(4-x^2)^2)",3)
f <- function(x){1.114283*exp(-(4-x^2)^2)}
plot(seq(-3,3,0.01),f(seq(-3,3,0.01)),lwd=2,lty=3,col="blue",type="l");lines(density(x,bw=0.05),lwd=2,lty=2,col="red")

#Mixed normal distribution
x <- rASS(500,2,"0.2/sqrt(2*pi)*exp(-x^2/2)+0.8/sqrt(2*pi*9)*exp(-(x-3)^2/2/9)",0.2)
f <- function(x){0.2/sqrt(2*pi)*exp(-x^2/2)+0.8/sqrt(2*pi*9)*exp(-(x-3)^2/2/9)}
plot(seq(-10,15,0.01),f(seq(-10,15,0.01)),lwd=2,lty=3,col="blue",type="l")
lines(density(x,bw=0.5),lwd=2,lty=2,col="red")

#Mixed gamma distribution
x <- rASS(500,6,"0.3*2^8/gamma(8)*x^7*exp(-2*x)+0.7*5^4/gamma(4)*x^3*exp(-5*x)",0.3)
f <- function(x){0.3*2^8/gamma(8)*x^7*exp(-2*x)+0.7*5^4/gamma(4)*x^3*exp(-5*x)}
plot(seq(0,8,0.01),f(seq(0,8,0.01)),lwd=2,lty=3,col="blue",type="l")
lines(density(x,bw=0.2),lwd=2,lty=2,col="red")

teacher7738/AdapSamp documentation built on May 4, 2019, 4:22 p.m.