Data Science Revealed: With Feature Engineering, Data Visualization, Pipeline Development, and Hyperparameter Tuning

© The Author(s), under exclusive license to APress Media, LLC, part of Springer Nature 2021

T. C. NokeriData Science Revealedhttps://doi.org/10.1007/978-1-4842-6870-4_2

2. Advanced Parametric Methods

Tshepo Chris Nokeri¹  

(1)

Pretoria, South Africa

 

The method covered in the previous chapter violated certain regression assumptions. It cannot capture noise, and as a result, it makes mistakes when predicting future instances. The most convenient way of combating this problem involves adding a penalty term to the equation.

This chapter introduces the novel concept of bias-variance trade-off, and it then covers regularized models like ridge regression, ridge with built-in cross-validation regression, and lasso regression. Last, it compares the performance ...