Extrapolate t-test based binning to a new data

Share:

Description

Extrapolates the binning of categorical variables to the new datasets.

Usage

1
apply.tbin(idv, train.output, test)

Arguments

idv

Predictor variables in the dataframe which are categorical and need to be binned.

train.output

The output object of the tbin function.

test

A new data set on which binning has to be extrapolated.

Details

This function performs binning on the new dataset based on the output object from the tbin function.

Value

Returns a dataframe which contains the extrapolated variables of the output object from tbin function appended to the new dataset.

Warning

New level error is thrown if the new dataset contains new levels other than what is present in the old dataset.

Author(s)

Mohan Manivannan

See Also

tbin, ctoc, apply.ctoc.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
train = as.data.frame(cbind(runif(1000, 10, 1000),sample(1:40, 1000,TRUE)))
colnames(train) = c("response","state")
train$state = as.factor(train$state)
train.output = tbin(dv = "response",idv = c("state"),train,25,TRUE)

# extrapolating the tbin function to a new dataset using apply.tbin
test = as.data.frame(sample(1:40, 100,TRUE))
colnames(test) = c("state")
test$state = as.factor(test$state)
test.output = apply.tbin(idv = c("state"), train.output, test)