predict.multistep: Multidimensional step function evaluation

In liso: Fitting lasso penalised additive isotone models

Description

Evaluates a multistep type function at a given value

Usage

## S3 method for class 'multistep'
predict(object, newx, ...)
## S3 method for class 'multistep'
e1 * e2

Arguments

object	A multistep object.
newx	Values to evaluate the represented function at. Each row is considered to be a separate observation.
...	Additional arguments for compatibility.
e1	Either a multistep object or a matrix to evaluate it at.
e2	Either a multistep object or a matrix to evaluate it at. One of e1, e2 must be a matrix, or vector.

Value

Produces a vector of results.

Note

predict(object, newx) is equivalent to object * newx, which is also equivalent to newx * object.

Author(s)

Zhou Fang

See Also

multistep

Examples

## Produces a 2d step function

set.seed(79)
n <- 100; p <- 2

## Choose some random knots
x <- matrix(runif(n * p, min = -2.5, max = 2.5), nrow = n, ncol = p)
obj = multistep(rep(0.1, (n-1)*p), x)
x2 <-  matrix(runif(n * p, min = -2.5, max = 2.5), nrow = n, ncol = p)
predict(obj,x) - obj*x

liso documentation built on May 29, 2017, 6:47 p.m.