mfa_data_reweight: Reweigh dataset based on MFA results

In HanjoStudy/quotidieR: Everyday functions for my personal use in analytics

Description

uses the results from the multiple factor analysis to reweigh all data based on sample results

Usage

mfa_data_reweight(dataset, mfa_res, index, dim, plot = T)

Arguments

dataset	the dataset that needs to be reweighted
mfa_res	a MFA object that contains the dimension reduction results
index	if index is provided it will be binded to dataframe in the end
dim	number of dimensions to keep
plot_scree	plots the eigenvalues against the factors for interpretation

Value

MFA Object

Examples

load_toolbox()

data(sales)

dataset <- sales
dataset[,1] <- factor(dataset[,1])

# Samle data and run MFA on the sample
sample <- sample_n(v, 1000)
 
load_pkg("FactoMineR")
mfa_res <- parMFA(sample,
                   group=c(1, 4, 12), 
                   type = c('n','s','s'), 
                   ncp=12, 
                   name.group=c('factor', 'purchases','seasonality'),
                   graph=FALSE, cores = 3)
                   
# The dimension needs to be specified
mfa_data_reweight(dataset, mfa_res)

mfa_data_reweight(dataset, mfa_res, dim = 12)
 

HanjoStudy/quotidieR documentation built on May 5, 2019, 6:13 p.m.