Curve fitting finds best function approximating data. Linear model: Y = a + bX (least squares). Non-linear models: exponential Y = ae^(bX), power Y = aX^b, logarithmic Y = a + b ln(X). Key concepts: choose model based on scatter plot pattern, minimize residual sum of squares. Common traps: over-fitting with complex models, extrapolating too far. Exam tips: plot data first, visually select appropriate model. Time-saving: transform non-linear to linear (log transformation), apply linear regression. Goodness of fit: R² indicates model quality (closer to 1 is better). Residual analysis: check for pattern violations. Applications: growth modeling, cost estimation, demand forecasting. Model assumptions: linear vs non-linear form choices. Understanding curve fitting essential for realistic modeling. Practice identifying and fitting different function types.