pred.simsl: SIMSL prediction function

In simsl: Single-Index Models with a Surface-Link

Description

This function makes predictions from an estimated SIMSL, given a (new) set of covariates. The function returns a set of predicted outcomes given the treatment values in a dense grid of treatment levels for each individual, and a recommended treatment level (assuming a larger value of the outcome is better).

Usage

pred.simsl(simsl.obj, newX = NULL, newA = NULL, newXm = NULL,
  single.index = NULL, L = 50, type = "link", maximize = TRUE)

Arguments

simsl.obj	a simsl object
newX	a (n-by-p) matrix of new values for the covariates X at which predictions are to be made.
newA	a (n-by-L) matrix of new values for the treatment A at which predictions are to be made.
newXm	a (n-by-q) matrix of new values for the covariates associated with the fitted main effect Xm at which predictions are to be made.
single.index	a length n vector specifying new values for the single-index at which predictions are to be made; the default is NULL.
L	when newA=NULL, a value specifying the length of the grid of A at which predictions are to be made.
type	the type of prediction required; the default "response" is on the scale of the response variable; the alternative "link" is on the scale of the linear predictors.
maximize	the default is TRUE, assuming a larger value of the outcome is better; if FALSE, a smaller value is assumed to be prefered.

Value

pred.new	a (n-by-L) matrix of predicted values; each column represents a treatment dose.
trt.rule	a (n-by-1) vector of suggested treatment assignments

Author(s)

Park, Petkova, Tarpey, Ogden

See Also

simsl,fit.simsl

simsl documentation built on Feb. 12, 2021, 5:17 p.m.