###################
#MI based on PCA (continuous)
###################

library(missMDA)
library(mice)


data(ozone)

## First the number of components has to be chosen 
nb <- estim_ncpPCA(ozone[,1:11])

## Multiple Imputation with Bayesian method
res.BayesMIPCA<-MIPCA(ozone[,1:11],ncp=nb$ncp,verbose=TRUE,method.mi = "Bayes")

## Diagnostics
res.over<-Overimpute(res.BayesMIPCA)

## Regression on the multiply imputed data set and pooling with mice
imp<-prelim(res.mi=res.BayesMIPCA,X=ozone[,1:11])#creating a mids object
fit <- with(data=imp,exp=lm(maxO3~T9+T12+T15+Ne9+Ne12+Ne15+Vx9+Vx12+Vx15+maxO3v))#analysis
res.pool<-pool(fit);summary(res.pool)#pooling

###################
#MI based on MCA (categorical)
###################

data(TitanicNA)

## First the number of components has to be chosen 
nb <- estim_ncpMCA(TitanicNA) ## Time-consuming, nb = 5

## Multiple Imputation
res.mi <- MIMCA(TitanicNA, ncp=5, verbose=TRUE)

## Analysis and pooling with mice
imp<-prelim(res.mi,TitanicNA)
fit <- with(data=imp,exp=glm(SURV~CLASS+AGE+SEX,family = "binomial"))
res.pool<-pool(fit)
summary(res.pool)