Time Series Forecasting: Prediction, Seasonality, and…

Article: Time Series – Prediction, Seasonality, Confidentiality

Time series forecasting is at the heart of finance, logistics, and energy.

Time series forecasting forms the basis for finance, logistics, and energy; key factors include seasonality, trends, and anomalies.

live demo · related simulation● LIVE

Classic Models: ARIMA, ETS, Prophet. ML: Gradient Boosting on Features

Classic models: ARIMA, ETS, Prophet. ML: Gradient boosting on lags and seasons, RNN/Temporal CNN/Transformers. Multiseries and hierarchical models.

Frequently asked questions

What is feature engineering: lags, moving statistics?

Feature engineering: lags, moving statistics, calendar features. Time splits, walk-forward validation. Aggregations at levels and reconciliation for hierarchies.

Is future information leakage incorrect?

Future information leakage, incorrect splits, regime shifts (changes in patterns), holidays/events.

Does high-quality forecasting require accurate splittings?

High-quality forecasting requires accurate splits, feature engineering, and adaptation to seasonal/event effects.

1) Decompose the series (trend/seasonal/res?

1) Decompose the series (trend/seasonal/residual). 2) Build a basic ARIMA/Prophet. 3) Add an ML model on engineered features. 4) Calibrate and verify stability.

▶ Try it live

Everything above runs in your browser — open Earthquake Wave Propagation Simulation and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.

▶ Open Earthquake Wave Propagation Simulation simulation

Time Series Forecasting: Prediction, Seasonality, and Confidence

Article: Time Series – Prediction, Seasonality, Confidentiality

Time series forecasting is at the heart of finance, logistics, and energy.

Classic Models: ARIMA, ETS, Prophet. ML: Gradient Boosting on Features

Frequently asked questions

▶ Try it live

What did you find?