03 — ML and DL Forecasting

Machine learning and deep learning approaches to time series forecasting, treating temporal prediction as a tabular supervised learning problem or a sequence modelling problem.

Notes

• ML and DL Forecasting — lag features for tree models, LSTMs, temporal convolutional networks, N-BEATS, temporal fusion transformers
• ML Forecasting — Implementation — LightGBM lag features, LSTM windowed dataset, walk-forward validation code

Links

← 02 — State-Space and Probabilistic → 06 — Training and Regularization

• Sequence Models — LSTM, GRU
• Tree-Based Models — gradient boosting for tabular forecasting