Backtest

Time series foundation models: forecasting tomorrow's electricity consumption for 3,517 commercial buildings

8 November 2026·Last updated: 9 November 2026·3412 words

Time Series Forecasting Foundation Models Chronos-2 LightGBM Energy Backtest Conformal Prediction Dagster MLflow Python

The TSFM tech watch applied to a fictional balance responsible party, on real public data: LightGBM, Chronos-2 and their blend against RTE’s day-ahead forecast, published and debiased. A 2.7% MAPE on the portfolio against 5.6% without a model, a router that does not beat the blend, a cold start handled by the naive forecast, Timestral and TimesFM 3.0 in an addendum, a Dagster service and its replayed monitoring.

Time series foundation models: zero-shot against the specialist

6 November 2026·5984 words·29 mins

Tech Watch Time Series Forecasting Foundation Models Chronos Zero-Shot Backtest MASE Prediction Intervals Conformal Prediction Statsforecast Python

When does a foundation model used without any training beat a model fitted on the series? What I observed over a year of French electricity load, with interval coverage and not only the mean error, then rebuilding a router between the two on 50 series.