Ridge Regression

Ridge regression is a technique that adds a penalty equivalent to the square of the magnitude of coefficients to the loss function. Unlike Lasso, it shrinks coefficients toward zero but does not set them exactly to zero.

This is useful when all input features are expected to contribute to the prediction, as it prevents any single feature from dominating the model. In crypto-finance, it is used to manage multicollinearity between different market indicators.

It provides a more stable model when features are highly correlated, which is common in financial time series.