Learn R Programming

PathSelectMP (version 1.1)

Backwards Variable Selection for Paths using M Plus

Description

Primarily for use with datasets containing only categorical variables, although continuous variables may be included as independent variables in paths. Using M Plus, backward variable selection is performed on all Total, Total Indirect, and then Direct effects until none of these effects have p-values greater than the specified target p-value. If there are missing values in the data, imputations are performed using the Mice package. Then selection is performed with the imputed data sets, and results are averaged.

Copy Link

Version

Install

install.packages('PathSelectMP')

Monthly Downloads

20

Version

1.1

License

GPL (>= 2)

Maintainer

William Terry

Last Published

December 19th, 2017

Functions in PathSelectMP (1.1)

CheckVarNames

Variable Name Check
AverageRRs

Calculate Average Risk Ratios over all imputed data sets
ConvertData2

AllBackwardSelect

Run Backward Variable Selection in Path Analysis with M Plus
AddOnAllInd

Indirect Effect Writer
AllSummary

Direct Effects Path Summaries
AddOnINDStatements

Indirect Effect List Creator
AllSummary2

Direct Effect Summary
CalculatRiskRatios

Calculate Risk Ratios from Direct Effects
CreateInitializeMatrix

Create Possible Path Matrix
AllTotalEffOutput

Summary of Total Effects
CreateSummaryMats

Extract and Summarize Direct Effects
MPlusBackwardSelect

Backward Selection Helper Function
ConvertData3

Used to make multiple list names for imputations
CreateTotalSummary

Summarizes and Averages Total Effects
ParseTotalEffects2

Parse and Exctract Total Effects
Initialize

Create Initial Objects
PathNames

M Plus Path description
CreateSummaryMats2

helper function for AddOnAllInd
DatImputations

Impute Data
WriteInitialInpFile

Write Inp File
NumCat

Number of Categories
mod

Simple Modular arithmetic
NumEndFile

Extract Number From INP and OUT Files
CreateTotalEffMat

Summarizes Total Effects
CountImputedEffect

Count and Average Effects for All Imputed Datasets
OnlyNumberElement

indexes list of lists
NewBinseqWrap

Possible Combinations of Elements, 1 from each list
ParseTotalEffects

Parse Total, Direct, and Indirect Effects
NewNamesThresh

Threshold Names
SpecialMatch

Match and delete elements
TotalRiskRatios

Calculate Risk Ratios for total effects
PathSelectMP-package

Backwards Variable Selection for paths using M Plus
Simulate

Simulate data set