GenerateMixData: Generates simulated data according to a specified INAR...

In inarmix: Mixture models for longitudinal count data.

Description

This function generates data according to a user-specified INAR mixture model. It returns a data frame which can be used immediately by inarmix.

Usage

GenerateMixData(m, coefs, autocorr, scale=NULL, mix, design.mat,
                return.labels=F,poisson=F)

Arguments

m	a numeric scalar (the number of subjects)
coefs	a numeric vector
autocorr	a numeric scalar
scale	a numerica scalar
mix	a numeric vector
design.mat	a numeric matrix (the design matrix)
return.labels	an indicator of whether or not to return the class labels
poisson	an indicator of whether or not the data should have marginal Poisson distributions.

Value

A data frame which contains the response and a subject identifier. The other columns contain the data entered from the design.mat argument.

Author(s)

Nicholas Henderson

Examples

XX <- cbind(rep(1,9),c(0:8)/4)
colnames(XX) <- c("const","time")
coefs <- rbind(c(-.2,0),c(1.2,.3))
autocorr <- c(.2,.2)
scale <- c(1.5,1.5)
mix.prop <- c(.8,.2)

testdat <- GenerateMixData(1000,coefs,autocorr,scale,mix.prop,XX)

Example output

Loading required package: Matrix

inarmix documentation built on May 2, 2019, 5:11 a.m.