Regression introductions fit line minimizing prediction error. Simple linear: ŷ = a + bx. Multiple regression: ŷ = a + b₁x₁ + b₂x₂ + ... Assumptions: (1) Linear relationship; (2) Independent observations; (3) Constant variance (homoscedasticity); (4) Normally distributed errors; (5) No multicollinearity (multiple regression). Assessing fit: R² (variance explained); residual plot (errors randomly scattered). Example: Predicting salary from education and experience. ŷ = 20000 + 5000×years_edu + 3000×years_exp. For 16 years education, 5 years experience: Predicted salary = 20000 + 80000 + 15000 = 115000. Exam tip: Verify assumptions before regression. Understand model limitations. Practice: Regression problems with interpretation.