predict.densEstBayes {densEstBayes}

Obtain the Bayesian density estimate from a densEstBayes() fit at new abscissae

Description

Values of the estimated density function, obtained via densEstBayes, are computed for new abscissae.

Usage

## S3 method for class 'densEstBayes'
predict(object,newdata,cred.fit = FALSE,credLev = 0.95,...)

Arguments

object

A densEstBayes() fit object.

newdata

A numerical vector of abscissae values.

cred.fit

Boolean flag:
TRUE = compute the lower and upper limits of (100*credLev)% pointwise credible intervals at 'newdata',
FALSE = do not computer credible interval limits (the default).

credLev

number between 0 and 1 such that the credible interval band has (100*credLev)% approximate coverage. The default value is 0.95.

...

A place-holder for other prediction parameters.

Value

A list is returned with the following components:

fit

numerical vector of ordinate values corresponding to the density estimate.

credLow.fit

numerical vector of ordinate values corresponding to the lower limits of the pointwise approximate (100*credLev)% credible set variability band.

credUpp.fit

numerical vector of ordinate values corresponding to the upper limits of the pointwise approximate (100*credLev)% credible set variability band.

Author(s)

Matt P. Wand [e-mail]

Examples

library(densEstBayes) ; data(OldFaithful2011)

# Obtain a density estimate for the `OldFaithful2011' data:

dest <- densEstBayes(OldFaithful2011,method = "SMFVB")

# Plot the density estimate:

plot(dest,xlab = "time interval between geyser eruptions (minutes)")
rug(jitter(OldFaithful2011,amount = 0.2),col = "dodgerblue")

# Obtain predictions at 60,70,80,90,100 and 110 seconds and
# add to them plot:

newdataVec <- seq(60,110,by = 10)
predictObj <- predict(dest,newdata = newdataVec,cred.fit = TRUE)
print(predictObj$fit)
points(newdataVec,predictObj$fit,col = "blue")

# Print and add to the plot the lower and upper limits of 
# the pointwise 95% credible intervals:

print(predictObj$credLow.fit)
print(predictObj$credUpp.fit)
points(newdataVec,predictObj$credLow.fit,col = "red")
points(newdataVec,predictObj$credUpp.fit,col = "red")