Time Series Forecasting

When working with Time Series data, the main goal is often predicting future values based on historical patterns. There are two major approaches:

Statistical Methods

Traditional models that explicitly capture trend, seasonality, and autocorrelation in time series data.

Common Techniques:

• Exponential Smoothing Forecasting
• ARIMA – Autoregressive Integrated Moving Average.
• SARIMA – Seasonal ARIMA for multiple seasonal patterns.
• Prophet – Handles seasonality and holidays well.

Implementations:

• Forecasting_Baseline.py – Naive or simple baseline forecasting.
• Forecasting_AutoArima.py – Automated ARIMA model selection.
• See: Forecasting_Exponential_Smoothing.py

Machine Learning Methods

Modern approaches that use feature-based forecasting or global models across multiple series. These methods require Feature Engineering for Time Series, such as lag features, rolling windows, and time-based features (hour, day, week). Possible Data Leakage if not setup correctly.

Examples:

• Random Forest for time series (global approach)
  • In ML_Tools see: https://github.com/rhyslwells/ML_Tools/blob/main/Explorations/Build/TimeSeries/TS_Random_Forest.py
• XGBoost
• LightGBM

Model Selection & Evaluation

To know if a forecasting model is good:

• Use proper Evaluation Metrics such as MAE, RMSE, MAPE.
• Apply time series cross-validation (rolling or expanding windows). See https://medium.com/@soumyachess1496/cross-validation-in-time-series-566ae4981ce4
• Time series cross validation techniques like Nested Cross Validation, Time Series Split Cross Validation, Blocked Cross Validation
• In ML_Tools: See https://github.com/rhyslwells/ML_Tools/blob/main/Explorations/Build/TimeSeries/TS_Cross_Validation.py

Resources: Time Series Forecasting Guide