What is the primary reason for applying a Yeo-Johnson transformation instead of a Box-Cox transformation?

Question

Data Science — Hard

What is the primary reason for applying a Yeo-Johnson transformation instead of a Box-Cox transformation?

Accepted Answer

The Box-Cox transformation is a power transformation that requires the input data to be strictly positive. The Yeo-Johnson transformation is an extension that supports zero and negative values, making it more versatile for real-world datasets where features may contain negative numbers or zeros. Both aim to stabilize variance and make data more Gaussian-like.