Linear and Logistic Regression

Key concepts:

• Linear regression: OLS, assumptions (linearity, homoscedasticity, independence, normality), multicollinearity.
• Logistic regression: link function, odds/odds ratio, decision thresholds, class imbalance handling.
• Regularization: L1 vs. L2, feature selection, interpretability.

Common interview checks:

• When does logistic regression outperform tree models?
• Explain coefficient meaning for both models.
• Diagnose overfitting and poor calibration.