Supplement for ridge and LASSO regression by kumar-navya · Pull Request #144 · DataScienceSpecialization/DataScienceSpecialization.github.io

In the lecture on Regularized Regression under the Practical Machine Learning course of Coursera's Data Science Specialization, we were introduced to the theoretical concepts of two penalized regression models: ridge and LASSO (Least Absolute Shrinkage and Selection Operator).

This is an attempt to:

1. Support that theory with a practical example using the mtcars dataset and the caret package to obtain a visual understanding of the concept of shrinking coefficients.

2. Compare goodness of fit on training data and prediction accuracy on test data across linear model (LM), ridge, and LASSO.

3. Explore the goodness of fit and prediction accuracy implications of feature selection in LM using LASSO.

github repo with content for ridge and LASSO