MMDai (version 0.1.0)

generateData: Generate random mixture multinomial dataset

Description

This function is used to generate random mixture multinomial dataset. It helps to evaluate our estimatior.

Usage

generateData(n, p, d, m, k, theta = rdirichlet(1, rep(1, k)), psi = array(rdirichlet(k * p, rep(1, d)), dim = c(k, p, d)), miss = 0)

Arguments

n

- number of observations

p

- number of variables

d

- number of categories

m

- number of trials

k

- number of components

theta

- a vector that sum to 1, denotes probability of latent components. Default value is Dirichlet random number with equal weight.

psi

- an array with dimension c(k,p,d), specific probability for each variables in each component. Default value is Dirichlet random number with equal weight.

miss

- rate of missing data in all observations. Missing mechanism is Missing Completely At Random (MCAR). The value of miss is from 0 to 1. Default value is 0. Missing data is denoted by NA. Here assume all trials in data[i,j,] are missing or observed at the same time.

Value

data - random dataset generated.theta - values of theta are used to generate random dataset.psi - values of psi are used to generate random dataset.

Examples

n<-100; k<-2; d<-2; m<-10; p<-2;
completedata<-generateData(n,p,d,m,k)
incompletedata<-generateData(n,p,d,m,k,miss = 0.25)