Nothing
Check.Assum <- function(Stage.1.Model){
# If no assumptions tested in Stage 1, assume that normality and homoscedasticity hold by default
if (Stage.1.Model$HomoNorm$Test.Assumptions==FALSE){Assume.Normality.S2 <- Assume.Homo.S2 <- TRUE}
# Alternative: use Levene/Breusch-pagan and Shapiro-Wilk tests
if (Stage.1.Model$HomoNorm$Test.Assumptions==TRUE){
# Check homoscedasticity
# Only conduct checks if there are predictors included in model, not relevant for intercept-only model
if (Stage.1.Model$HomoNorm$Num.Preds > 1){
if (Stage.1.Model$HomoNorm$Contains.Numeric.Pred==FALSE){ # if no continuous predictors, use levene test homo
ifelse(test = Stage.1.Model$HomoNorm$Levene$`Pr(>F)`[1] <= Stage.1.Model$HomoNorm$Alpha.Levene, yes = Assume.Homo.S2<-FALSE,
no = Assume.Homo.S2<-TRUE)}
if (Stage.1.Model$HomoNorm$Contains.Numeric.Pred==TRUE){ # if continuous predictors, use Breusch-Pagan test homo
ifelse(test = Stage.1.Model$HomoNorm$Breusch.Pagan$p.value <= Stage.1.Model$HomoNorm$Alpha.BP, yes = Assume.Homo.S2<-FALSE,
no = Assume.Homo.S2<-TRUE)}
}
# If no predictors included in model, irrelevant
if (Stage.1.Model$HomoNorm$Num.Preds ==1){
Assume.Homo.S2<-TRUE #LS
}
# Check normality
if (Assume.Homo.S2==TRUE){ # for delta computed under assumption homoscedasticity
ifelse(test = Stage.1.Model$HomoNorm$Shapiro.Wilk$p.value <= Stage.1.Model$HomoNorm$Alpha.Shapiro, yes = Assume.Normality.S2<-FALSE,
no = Assume.Normality.S2<-TRUE)
}
if (Assume.Homo.S2==FALSE){ # for delta computed not assuming homoscedasticity
ifelse(test = Stage.1.Model$NoHomoNorm$Shapiro.Wilk$p.value <= Stage.1.Model$NoHomoNorm$Alpha.Shapiro, yes = Assume.Normality.S2<-FALSE,
no = Assume.Normality.S2<-TRUE)
}
}
fit <- list(Assume.Homo.S2=Assume.Homo.S2, Assume.Normality.S2=Assume.Normality.S2,
Call=match.call())
class(fit) <- "Check.Assum.S2"
fit
}
Any scripts or data that you put into this service are public.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.