factorMerger: A Set of Tools to Support Adaptive Post-Hoc Fusing of Groups

The aim of this project is to create an algorithm of post-hoc testing that would enable to extract hierarchical structure of factors with respect to a given response.

Installing and loading factorMerger

factorMerger can be installed from CRAN as follows:

install.packages("factorMerger")

To install and load the latest version offactorMerger from Github run:

devtools::install_github("MI2DataLab/factorMerger", build_vignettes = FALSE)

Materials for factorMerger

A longer description of factorMerger may be found here (https://arxiv.org/abs/1709.04412).

Cheatsheet, the pdf version

Working with factorMerger

Examples

Survival analysis

library(factorMerger)
library(forcats) # distinguish meaningful factors (fct_lump)

data("BRCA")
brcaSurv <- survival::Surv(time = BRCA$time, event = BRCA$vitalStatus)
drugName <- fct_lump(as.factor(BRCA$drugName), prop = 0.05) 

drugNameFM <- mergeFactors(response = brcaSurv[!is.na(drugName)], 
                           factor = drugName[!is.na(drugName)], 
                           family = "survival")

plot(drugNameFM, nodesSpacing = "effects", gicPanelColor = "grey2")