build.logistic.model: Build a logistic model

In ProfessionalFarmer/loonR: Common function in bioinformatics analysis

Build a logistic model

Description

Build a logistic model

Usage

build.logistic.model(
  df,
  group,
  seed = 666,
  scale = TRUE,
  direction = "backward",
  rms = FALSE
)

Arguments

df	Column is gene/miRNA, Row is sample
group	Two levels. Second unique variable is defined as experiment group
direction	backward c("both", "backward", "forward")
rms	If TRUE, use rms instead of glm to build the model. Useful for validation and calibration function in rms package.

Value

A list. list(model=glm.fit, StepwiseModel=elimination, eliminationCandidates=stringr::str_remove_all(names(unclass(coef(elimination))-c(1)),''') )

Examples

data("LIRI")
lg.res <- loonR::build.logistic.model(LIRI[,3:5],LIRI$status)
lg.res

ProfessionalFarmer/loonR documentation built on Oct. 9, 2024, 9:56 p.m.