Equi-width binning

 

The range of variable value is divided into k equal width intervals, each of which is regarded as a bin. Here, only boundary is considered, and the sample size in each bin may be different.

For example, Equi-width binning the “Fare” variable


A

1

=file("D://titanic.csv").import@qtc()

2

=A1.(Fare).max()

3

=A1.(Fare).min()

4

3

5

=(A2-A3)/A4

6

=A4.((~-1)*A5+A3)

7

=A1.derive(if(Fare<=A6(2),"low",if(Fare>A6(3),"hign","middle")):Fare_equiwidth_binning)

A5 Calculate the length of each bin

A6 Calculate the boundary values for each bin

A7 Binning fare according to boundary values