Movatterモバイル変換

# StepReg

• #### An Rpackage for stepwise regression analysis

StepReg is an R package that streamlines stepwise regression analysisby supporting multiple regression types, incorporating popular selectionstrategies, and offering essential metrics.

Key Features

• Multiple Regression Types: Linear, logistic, Cox,Poisson, Gamma, and negative binomial regression
• Selection Strategies: Forward selection, backwardelimination, bidirectional elimination, and best subsets
• Selection Metrics: AIC, AICc, BIC, CP, HQ, adjRsq,SL, SBC, IC(3/2), IC(1)
• Advanced Features:
  • Strata variables for Cox regression
  • Continuous-nested-within-class effects
  • multivariable multiple linear stepwise regression
• Multicollinearity Detection: Automatic detectionand handling of multicollinearity
• Visualization: Plot functions for variableselection processes
• Reporting: Export results in various formats (HTML,DOCX, XLSX, PPTX)
• Shiny App: Interactive web interface fornon-programmers

Installation

Install from CRAN

pak::pkg_install("StepReg")

or

install.packages("StepReg")

Or install from GitHub

devtools::install_github("JunhuiLi1017/StepReg")

Quick Start

library(StepReg)# Basic linear regressiondata(mtcars)formula<- mpg~ .res<-stepwise(formula = formula,data = mtcars,type ="linear",strategy ="bidirection",metric ="AIC")# View resultsressummary(res$bidirection$AIC)

Advanced Features

Strata Variables in CoxRegression

library(survival)data(lung)lung$sex<-factor(lung$sex)# Cox regression with strataformula<-Surv(time, status)~ age+ sex+ ph.ecog+strata(inst)res<-stepwise(formula = formula,data = lung,type ="cox",strategy ="forward",metric ="AIC")

data(mtcars)mtcars$am<-factor(mtcars$am)# Nested effectsformula<- mpg~ am+ wt:am+ disp:am+ hp:amres<-stepwise(formula = formula,data = mtcars,type ="linear",strategy ="bidirection",metric ="AIC")

citation("StepReg")