Lightning ⚡️ fast forecasting with statistical and econometric models.
-
Updated
Oct 12, 2026 - Python
Lightning ⚡️ fast forecasting with statistical and econometric models.
Forecast time series with >95% accuracy
An end-to-end 48-hour operational forecasting pipeline for hourly energy grid demand benchmarking MSTL + ARIMA against Amazon Chronos-Mini Transformers
Fork of Nixtla/statsforecast. Lightning ⚡️ fast forecasting with statistical and econometric models.
A three stage architecture for forecasting New Zealand's residential construction activity (consents, gross floor area, value) at typology resolution. It uses MSTL, LSTMs for residual learning, and empirical minimum trace reconciliation (MinT). Code for Christoforatos & Pickering (2026), ESWA.
A Python package for forecasting integer counts in stationary time series using MSTL, Prophet, and SARIMA with conditional seasonalities.
To associate your repository with the mstl topic, visit your repo's landing page and select "manage topics."