Skip to content
varun.rout

Jump to

Search pages, projects and repos

← Back to work

Energy markets · production

ProfessionalReal data

Residual-Load Forecasting & Price Curves

Quantitative Market Analyst · E.ON Energy Markets · 2025

What this is

Rebuilt hourly power forecasting around a residual-load framework (demand net of wind and solar), benchmarked against the previous demand-based methodology and monitored for drift in production.

Metrics

Forward curve

+15% far / +9.7% near

vs previous methodology

Christmas fix

RMSE -18%

vs pre-fix

Signal

Strong residual-demand/price relationship

Stack

Azure ML StudioAzure DevOpsPythonPower BIDrift detectionTime-series

Provenance

Professional work at E.ON Energy Markets. Methodology and relative improvements only: no E.ON data, code, models, absolute error figures or commercially sensitive parameters are published here.

What I would not claim

  • No absolute forecast error, volume or price figures are disclosed.
  • Improvements are relative to an internal baseline that is deliberately not characterised.
  • No E.ON code, data or model artefacts are shared; none of this work is reproducible from this site.

Explore the deltas

The same before/after figures, interactive. Toggle between the three roles.

Residual-load framework improved forward-curve accuracy on both near and far seasons; the Christmas-period fix followed a holiday-specific error diagnosis. Relative improvements only, against an internal baseline.