Ridge regression adds a penalty term (L2 regularization) to the linear regression cost function. It helps prevent overfitting by forcing the coefficients to be small.

👉 What it does: Adds a penalty to shrink coefficients (but never makes them zero)

✅ When to use:

• You have many features
• Features are highly correlated
• You want to reduce overfitting, but keep all variables

🧠 Real-life examples:

• Predicting stock prices using dozens of financial indicators
• Modeling customer behavior with many overlapping variables (age, income, spending habits)
• Healthcare: predicting outcomes using many correlated biomarkers

💡 Key idea: 👉 Keeps all features, just reduces their impact