SpatialExtremes (version 1.2-0)

predict: Prediction of smoothing spline with radial basis functions

Description

This function predicts the response from a fitted smoothing spline.

Usage

## S3 method for class 'pspline':
predict(object, newdata, \dots)

Arguments

object

An object of class ``pspline''. Most often, it will be the output of the function rbpspline.

newdata

An optional data frame in which to look for variables with which to predict. If omitted, the fitted values are used.

...

further arguments passed to or from other methods.

Value

• 'predict.pspline' produces a vector of predictions or a matrix of predictions.

See Also

predict

Examples

## 1- Define a function to approximate
fun <- function(x)
  sin(3 * pi * x)

## 2- Compute the response from fun - and adding a noise
x <- seq(0, 1, length = 200)
y <- fun(x) + rnorm(200, 0, sqrt(0.4))

## 2- Fit a penalized smoothing spline
n.knots <- 30
knots <- quantile(x, prob = 1:n.knots / (n.knots + 2))
fitted <- rbpspline(y, x, knots, degree = 3)

## 3- Prediction from the fitted spline
plot(x, y, pch = 2, col = "red")
plot(fun, from = 0, to = 1, add = TRUE)
pred <- predict(fitted)
lines(pred[,1], pred[,2], col = "blue", pch = 3)