apply.ctoc: Applying Categorical to Continuous conversion to a new...

In corkscrew: Preprocessor for Data Modeling

Description

Extrapolating the categorical to continuous conversion that is calculated from one dataframe to another dataframe.

Usage

apply.ctoc(y, x, data, newdata, min.obs)

Arguments

y	Response variable (categorical or continuous).
x	Predictor variables in the dataframe which are categorical and need to be converted into continuous.
data	Name of the dataframe from which the values of the categories have to be calculated.
newdata	Name of the dataframe to which the values of the categories have to be applied.
min.obs	The minimum number of observations within a category in a categorical variable to get converted into a continuous feature. All the categories which have observations less than the min.obs will form a different category.

Details

This function is only for categorical variables. The min.obs refers to the minimum number of observations in the "data".

Value

Returns a dataframe with converted features without replacing the original ones.

Author(s)

Santhosh Sasanapuri

See Also

ctoc, tbin, apply.tbin.

Examples

data(ChickWeight)
set.seed(2)
sample_ex <- sample(nrow(ChickWeight), size = 289, replace = FALSE, prob = NULL)
train <- ChickWeight[sample_ex,]
test <- ChickWeight[-sample_ex,colnames(ChickWeight) != "weight"]
# Returns the test dataframe with an added column "Diet_cont" by extrapolating it from train
head(apply.ctoc(y = "weight", "Diet", data = train, newdata = test, min.obs = 60))

Example output

Loading required package: ggplot2
Loading required package: gplots

Attaching package: 'gplots'

The following object is masked from 'package:stats':

    lowess

Loading required package: RColorBrewer
Loading required package: igraph

Attaching package: 'igraph'

The following objects are masked from 'package:stats':

    decompose, spectrum

The following object is masked from 'package:base':

    union

  Diet Time Chick Diet_cont
1    1    0     1  102.4865
2    1    2     1  102.4865
3    1    4     1  102.4865
4    1   16     1  102.4865
5    1   18     1  102.4865
6    1    4     2  102.4865

corkscrew documentation built on May 1, 2019, 8:23 p.m.