Learn R Programming

lmSupport (version 2.9.13)

modelAssumptions: Assess Linear Model Assumptions

Description

Provides diagnositic graphs and score tests to evaluate linear model assumptions of normality, constant variance and linearity. Follows best practices and uses many functions from car package.

Usage

modelAssumptions(Model, Type = "NORMAL", ID=row.names(Model$model), one.page = TRUE)

Arguments

Model

a linear model produced by lm.

Type

Type =c('NORMAL', 'CONSTANT', 'LINEAR') for normally distributed residuals with constant variance, and linear (e.g., mean of residuals 0 for all Y')

ID

Use to identify points. Default = row.names(model$model). NULL = no identification

one.page

logical; display all graphs on one page if TRUE (Default).

References

Fox, J. (1991). Regression diagnostics. SAGE Series (79) Quantitative Applictions in the Social Science.

Examples

Run this code
# NOT RUN {
data(BAC)
m = lm(FPS~BAC+TA, data=BAC)
modelAssumptions(m,'NORMAL')
modelAssumptions(m,'CONSTANT')
modelAssumptions(m,'LINEAR', ID=NULL)
# }

Run the code above in your browser using DataLab