Time-series & forecasting¶
Time-series data — prices, metrics, sensor readings, demand — has temporal structure that ordinary ML ignores at its peril. This chapter covers stationarity, autocorrelation, classical models (ARIMA, exponential smoothing), volatility models (GARCH), the ML/DL forecasters that increasingly beat classical baselines, temporal feature engineering, and — crucially — the rigorous, leakage-free evaluation (walk-forward, purged/ combinatorial CV, forecast-accuracy metrics, Deflated Sharpe) that separates a real edge from backtest luck. …
🔒 La suite est en accès freemium — lecture complète 100 % gratuite
Tu lis ici l'aperçu libre. Le reste du chapitre (code, schémas, maths, exercices) fait partie du livre complet : crée un compte gratuit (30 secondes, aucun paiement) pour tout lire.