confband.pw: Pointwise Confidence Bands

In acid: Analysing Conditional Income Distributions

Description

This function computes pointwise confidence bands for samples of the presumed distribution of the parameter estimator.

Usage

confband.pw(samples, level = 0.95)

Arguments

samples	matrix containing samples of the presumed distribution of the parameter estimator.
level	the desired confidence level.

Value

lower	a vector containing the lower bound of the confidence band.
upper	a vector containing the lower bound of the confidence band.

Note

This function is mainly derived from the work of T. Krivobokova, T. Kneib and G. Claeskens.

Author(s)

Alexander Sohn

References

T. Krivobokova, T. Kneib, G. Claeskens (2010): Simultaneous Confidence Bands for Penalized Spline Estimators, in: Journal of the American Statistical Association, Vol. 105(490), pp.852-863.

Examples

mu<-1:20
n<-1000
mcmc<-matrix(NA,n,20)
for(i in 1:20){
  mcmc[,i]<- rnorm(n,mu[i],sqrt(i))
}

plot(mu,type="l",ylim=c(-10,30),lwd=3)
lines(confband.pw(mcmc)$lower,lty=2)
lines(confband.pw(mcmc)$upper,lty=2)
lines(confband.kneib(mcmc)$lower,lty=3)
lines(confband.kneib(mcmc)$upper,lty=3)

acid documentation built on May 1, 2019, 10:14 p.m.