nullModel

From caret v5.09-012 by Max Kuhn

Fit a simple, non-informative model

Fit a single mean or largest class model

Keywords: models

Usage

nullModel(x, ...)
## S3 method for class 'default':
nullModel(x = NULL, y, ...)
## S3 method for class 'nullModel':
predict(object, newdata = NULL, type = NULL, ...)

Arguments

x: An optional matrix or data frame of predictors. These values are not used in the model fit
y: A numeric vector (for regression) or factor (for classification) of outcomes
...: Optional arguments (not yet used)
object: An object of class nullModel
newdata: A matrix or data frame of predictors (only used to determine the number of predictions to return)
type: Either "raw" (for regression), "class" or "prob" (for classification)

Details

nullModel emulates other model building functions, but returns the simplest model possible given a training set: a single mean for numeric outcomes and the most prevalent class for factor outcomes. When class probabilities are requested, the percentage of the training set samples with the most prevalent class is returned.

Value

The output of nullModel is a list of class nullModel with elements
callthe function call
valuethe mean of y or the most prevalent class
levelswhen y is a factor, a vector of levels. NULL otherwise
pctwhen y is a factor, a data frame with a column for each class (NULL otherwise). The column for the most prevalent class has the proportion of the training samples with that class (the other columns are zero).
nthe number of elements in y
predict.nullModel returns a either a factor or numeric vector depending on the class of y. All predictions are always the same.

Examples

outcome <- factor(
                  sample(letters[1:2], 
                         size = 100, 
                         prob = c(.1, .9), 
                         replace = TRUE))
useless <- nullModel(y = outcome)
useless
predict(useless, matrix(NA, nrow = 10))

Documentation reproduced from package caret, version 5.09-012, License: GPL-2

Community examples

Looks like there are no examples yet.