modelCompare: F-tests for nested models

In lmSupport: Support for Linear Models

Description

Calculates F-test to compare two models to determine if ModelA significantly reduces SSE from ModelC. Also reports Partial eta2 and Delta R2 for this model comparison. ModelC should contain subset of ModelA regressors.

Usage

modelCompare(ModelC, ModelA)

Arguments

ModelC	a linear model, produced by lm. This compact model should include a subset of regressors from ModelA
ModelA	a linear model, produced by lm

. This augmented model should include all regressors from ModelC plus additional regressors.

Details

Calculates F test for model comparison F = ((sseC - sseA)/ (pA - pC)) / (sseA / (N-pA)) ndf = pA - pC ddf = N -P

Value

Returns a list with results for model comparison, sses, and other relevant fields

Author(s)

John J. Curtin [e-mail]

Examples

##NOT RUN
##mC = lm(FPS~BAC, data=BAC)
##mA = lm(FPS~BAC+TA, data=BAC)
##modelCompare(mC, mA)

Example output

Registered S3 methods overwritten by 'lme4':
  method                          from
  cooks.distance.influence.merMod car 
  influence.merMod                car 
  dfbeta.influence.merMod         car 
  dfbetas.influence.merMod        car 

lmSupport documentation built on May 2, 2019, 2:14 p.m.