The science of predicting the future from the past. Decomposition (trend, seasonality, residual), stationarity & differencing, autocorrelation (ACF & PACF), white noise & random walks, moving averages, exponential smoothing, Holt's linear trend, Holt-Winters seasonal, AR / MA / ARMA / ARIMA / SARIMA / ARIMAX, Box-Jenkins methodology, STL decomposition, forecast accuracy (MAE, RMSE, MAPE, MASE), time-series cross-validation, anomaly detection (STL residuals, isolation forest), Prophet & NeuralProphet, LSTM & Transformer models, multivariate time series & VAR, Granger causality, spectral & wavelet analysis, and real applications — sales & demand forecasting, energy demand, weather & climate, epidemic curves, finance (with caution), tools (Python statsmodels, R forecast/fable, Facebook Prophet), and a capstone forecasting project. 35 units, 525 lessons.