confpred: Conformal prediction

In lava: Latent Variable Models

Conformal prediction

Description

Conformal predicions using locally weighted conformal inference with a split-conformal algorithm

Usage

confpred(object, data, newdata = data, alpha = 0.05, mad, ...)

Arguments

object	Model object (lm, glm or similar with predict method) or formula (lm)
data	data.frame
newdata	New data.frame to make predictions for
alpha	Level of prediction interval
mad	Conditional model (formula) for the MAD (locally-weighted CP)
...	Additional arguments to lower level functions

Value

data.frame with fitted (fit), lower (lwr) and upper (upr) predictions bands.

Examples

set.seed(123)
n <- 200
x <- seq(0,6,length.out=n)
delta <- 3
ss <- exp(-1+1.5*cos((x-delta)))
ee <- rnorm(n,sd=ss)
y <- (x-delta)+3*cos(x+4.5-delta)+ee
d <- data.frame(y=y,x=x)

newd <- data.frame(x=seq(0,6,length.out=50))
cc <- confpred(lm(y~splines::ns(x,knots=c(1,3,5)),data=d), data=d, newdata=newd)
if (interactive()) {
plot(y~x,pch=16,col=lava::Col("black"),ylim=c(-10,10),xlab="X",ylab="Y")
with(cc,
     lava::confband(newd$x,lwr,upr,fit,
        lwd=3,polygon=TRUE,col=Col("blue"),border=FALSE))
}

lava documentation built on May 29, 2024, 3:10 a.m.