Compounding (version 1.0.2)

momentCompound: function momentCompound

Description

Function momentCompound calculates moments of the random variable X.

Usage

momentCompound(k,parent,compound,compoundDist,params,...)

Arguments

k

integer number representing moment's order

parent

name of the parent distribution. It can be any continuous distribution supported by R.

compound

name of the compound distribution. It can be any discrete distribution supported by this package.

compoundDist

list of available compounding distributions

params

Parameter or list of parameters of compounding distribution.

...

Parameters of continuous distribution could be provided as additional parameters.

Details

Parameters of the parent distribution must be provided in the same way as it is in built in R functions. See http://127.0.0.1:23174/library/stats/html/Distributions.html

References

S. Nadarajah, B.V. Popovic, M.M. Ristic (2012) Compounding: an R package for computing continuous distributions obtained by compounding a continuous and a discrete distribution, Computational Statistics, DOI 10.1007/s00180-012-0336-y, http://www.springerlink.com/content/6r464013w6mp3545/

Examples

compoundDist <- c("geometric","poisson","negativebinomial","binomial",
"logarithmic","binomialbinomial","binomialpoisson",
"poissonbinomial","neymantypea","polyaaeppli",
"poissonpascal","pascalpoisson",
"logarithmicbinomial","logarithmicpoisson",
"poissonlindley",
"hyperpoisson","yule","waring","kattitypeh1",
"kattitypeh2","neymantypeb","neymantypec",
"hypergeometric","thomas")
k<-3
parentD<-"beta"
compoundD<-"hypergeometric"
params<-c(3,2,0.5)
momentCompound(k,parentD,compoundD,compoundDist,params,shape1=2,shape2=0.3)

## The function is currently defined as
function(k, parent, compound, compoundDist,params, ...) {
    if (!exists(paste("p",parent,sep=""))) {
        return(paste("The parent distribution",parent,"doesn't exist"))
    }
    if (!is.element(compound,compoundDist)) {
        return(paste("The discrete distribution",compound,"doesn't exist"))
    }
    Finv <- get(paste("q", parent, sep = ""), mode = "function")
    phi <- get(paste("pgf",compound,sep=""), mode = "function")
    phiD <- get(paste("pgfD",compound,sep=""), mode = "function")
    fint <- function(x) phiD(1-x,params)*(Finv(x,...))^k/(1-phi(0,params))
    return(integrate(fint,lower=0,upper=1)$value)
}