Box-Cox transformation

 

Box-cox transformation is a generalized power transformation method proposed by Box and Cox in 1964. It is a kind of data transformation commonly used in statistical modeling for cases where continuous quantities do not satisfy normal distribution

The general form of the Box-Cox transformation is:

..

For example, a Box-Cox transformation is performed on the "SalePrice" in the house price data


A

1

=T("D://house_prices_train.csv")

2

0.5

3

=A1.derive(if(A2==0,ln(SalePrice),(power(SalePrice,A2)-1)/A2):Box_Cox_SalePrice)

A2 Input transformation parameter λ

A3 Transform the SalePrice