predict.ordPen: Predict method for ordPen objects

Description

Obtains predictions from an ordPen object.

Usage

# S3 method for ordPen
predict(object, newx, newu = NULL, newz = NULL,
  offset = rep(0,nrow(as.matrix(newx))), 
  type = c("link", "response", "class"), ...)

Value

A matrix of predictions whose columns correspond to the different values of the penalty parameter lambda of the ordPen object.

Arguments

object: an ordPen object.
newx: the matrix (or data.frame) of new observations of the considered ordinal predictors, with each column corresponding to one predictor and containing numeric values from {1,2,...}.
newu: a matrix (or data.frame) of new observations of the additional categorical (nominal) predictors, with each column corresponding to one (additional) predictor and containing numeric values {1,2,...}.
newz: a matrix (or data.frame) of new observations of the additional metric predictors, with each column corresponding to one (additional) predictor.
offset: potential offset values.
type: the type of prediction; type = "link" is on the scale of linear predictors, whereas type = "response" is on the scale of the response variable, i.e., type = "response" applies the inverse link function to the linear predictors. type = "class" is only available for cumulative logit models and returns the class number with the highest fitted probability.
...: additional arguments (not supported at this time).

Author

Jan Gertheiss, Aisouda Hoshiyar

Examples

Run this code

# the training data
set.seed(123)

# generate (ordinal) predictors
x1 <- sample(1:8,100,replace=TRUE)
x2 <- sample(1:6,100,replace=TRUE)
x3 <- sample(1:7,100,replace=TRUE)

# the response
y <- -1 + log(x1) + sin(3*(x2-1)/pi) + rnorm(100)

# x matrix
x <- cbind(x1,x2,x3)

# lambda values
lambda <- c(1000,500,200,100,50,30,20,10,1)

# selecting and/or smoothing/fusing
o1 <- ordSmooth(x = x, y = y, lambda = lambda)
o2 <- ordSelect(x = x, y = y, lambda = lambda)
o3 <- ordFusion(x = x, y = y, lambda = lambda)

# new data
x1 <- sample(1:8,10,replace=TRUE)
x2 <- sample(1:6,10,replace=TRUE)
x3 <- sample(1:7,10,replace=TRUE)
newx <- cbind(x1,x2,x3)

# prediction
round(predict(o1, newx), digits=3)
round(predict(o2, newx), digits=3)
round(predict(o3, newx), digits=3)