earth: predict.varmod – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

predict.varmod

Predict with a varmod model

Description

You probably won't need to call this function directly. It is called by predict.earth when that function's interval argument is used.

Usage

## S3 method for class 'varmod'
predict(
    object  = stop("no 'object' argument"),
    newdata = NULL,
    type    = c("pint", "cint", "se", "abs.residual"),
    level   = .95,
    trace   = FALSE,
    ...)

Arguments

`object`	A `varmod` object.
`newdata`	Make predictions using `newdata`. Default is NULL, meaning return values predicted from the training set.
`type`	Type of prediction. This is the `interval` argument of `predict.earth`. One of `"pint"` Prediction intervals. `"cint"` Confidence intervals. Cannot be used with `newdata`. `"se"` Standard error of the parent model residuals. `"abs.residual"` The absolute residuals of the parent model on which the residual model regresses.
`level`	Confidence level for the `interval` argument. Default is `.95`, meaning construct 95% confidence bands (estimate the 2.5% and 97.5% levels).
`trace`	Currently unused.
`...`	Unused, but provided for generic/method consistency.

Note

predict.varmod is called by predict.earth when its interval argument is used.

Examples

data(ozone1)

set.seed(1) # optional, for cross validation reproducibility

# note: should really use ncross=30 below but for a quick demo we don't

earth.mod <- earth(O3~temp, data=ozone1, nfold=10, ncross=3, varmod.method="lm")

# call predict.earth, which calls predict.varmod

predict(earth.mod, newdata=ozone1[200:203,], interval="pint", level=.95)

earth

Multivariate Adaptive Regression Splines

v5.3.0

GPL-3

Authors

Stephen Milborrow. Derived from mda:mars by Trevor Hastie and Rob Tibshirani. Uses Alan Miller's Fortran utilities with Thomas Lumley's leaps wrapper.

Initial release