plfm (version 2.2.6)

plot.stepplfm: Plot fit of stepplfm objects

Description

Plot method to visualize the fit of probabilistic latent feature analysis objects with different numbers of features.

Usage

# S3 method for stepplfm
plot(x,which="BIC",...)

Arguments

x

List of probabilistic latent feature analysis objects returned by stepplfm.

which

Fit criterion for which models with different numbers of features are compared. The argument which can take the following values: "AIC", "BIC", "Deviance", "Chisquare", "Correlation", "VAF"

...

Further arguments are ignored.

Examples

if (FALSE) {
## example 1:Perceptual analysis of associations between car models and car attributes

##load car data
data(car)


## compute 5 runs of disjunctive and conjunctive models with 1 up to 4 features
car.lst<-stepplfm(minF=1,maxF=4,maprule="disj/conj",freq1=car$freq1,
                  freqtot=car$freqtot,M=5)

## visualize the fit of models with different mapping rules 
## and a different number of features

par(pty="s")
par(mfrow=c(2,2))
plot(car.lst,which="BIC")
plot(car.lst,which="AIC")
plot(car.lst,which="VAF")

}

if (FALSE) {
## example 2: analysis on determinants of anger-related behavior

## load anger data
data(anger)

## compute 1 run of disjunctive and conjunctive models with 1 up to 3 features
anger.lst<-stepplfm(minF=1,maxF=3,maprule="disj/conj",freq1=anger$freq1,
                    freqtot=anger$freqtot,M=1)

## visualize the fit of models with different mapping rules 
## and a different number of features

par(pty="s")
par(mfrow=c(2,2))
plot(anger.lst,which="BIC")
plot(anger.lst,which="AIC")
plot(anger.lst,which="VAF")
}