https://gwils28.github.io/en/tags/forecasting/Recent content in Forecasting on Kwak JazzHugo -- gohugo.ioen© 2026 Wilson GomaMon, 09 Nov 2026 00:26:26 +0100https://gwils28.github.io/en/projects/flexgrid-prevision-j1/Sun, 08 Nov 2026 00:00:00 +0000https://gwils28.github.io/en/projects/flexgrid-prevision-j1/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.https://gwils28.github.io/en/posts/tsfm-zero-shot-specialiste/Fri, 06 Nov 2026 00:00:00 +0000https://gwils28.github.io/en/posts/tsfm-zero-shot-specialiste/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.https://gwils28.github.io/en/posts/impact-causal-series-temporelles/Tue, 27 Oct 2026 00:00:00 +0000https://gwils28.github.io/en/posts/impact-causal-series-temporelles/A counterfactual is a forecast. What I observed building one by hand, measuring its coverage on 100 panels with a known truth, planting four violations of the design, and applying it to Volkswagen’s share price after dieselgate.https://gwils28.github.io/en/posts/frequence-reentrainement/Sat, 17 Oct 2026 00:00:00 +0000https://gwils28.github.io/en/posts/frequence-reentrainement/Accuracy, freshness and stability don’t call for the same retraining frequency. What I observed on a ten-policy bench, an analytical model, the Monash electricity data and a causal simulation.https://gwils28.github.io/en/projects/w37-reconciliation-hierarchique/Thu, 08 Oct 2026 00:00:00 +0000https://gwils28.github.io/en/projects/w37-reconciliation-hierarchique/MinT rewritten from scratch, its assumptions tested one by one, and an éCO2mix mini-project (12 regions → France) where bottom-up beats the reference method, over 356 origins, two years and two base models.https://gwils28.github.io/en/posts/reconciliation-hierarchique/Tue, 06 Oct 2026 00:00:00 +0000https://gwils28.github.io/en/posts/reconciliation-hierarchique/MinT guarantees coherence, not accuracy, and not calibration. Shown through theory, simulation, French grid data and count sales.