Hourly Price Calendar

Wind Power Outlook

Solar Power Outlook

Forecast History

How the Forecast Works

The forecasts on this site are generated with an XGBoost-based machine learning pricing model (gradient-boosted decision trees) that estimates hourly electricity prices from weather, generation, and market features. Each hour is evaluated independently based on its expected conditions, which lets the model react to changes in wind, imports, generation availability, and seasonal demand.

The model is trained on historical market data starting from 2023. Weather data comes from ECMWF, using recent historical weather together with forward forecasts for each area.

Key Inputs Considered by the Model

• Area-specific wind and temperature sampled from ERA5 history and ECMWF forecasts
• Regional wind fields from neighboring countries
• Temperature forecasts capturing demand and seasonality
• Renewable generation availability (including nuclear where applicable)
• Cross-border transfer capacities for the forecast area
• Solar power output built from area solar forecasts and irradiance features
• Hydrology state features for Finland and Norway
• Calendar effects such as year, weekday, hour, and holidays

Wind Forecasting

Recent ENTSO-E actuals are used to anchor the wind picture. The raw wind-power model is built from area-specific weather features, historical production, and installed capacity. Finland remains a special case on the power side: Fingrid real-time data and Fingrid's operational wind-power forecast are used directly where available, with model inference filling remaining gaps.

Model Logic and Interpretation

The model learns statistical relationships between explanatory variables and price outcomes. High wind or mild temperatures are typically associated with lower prices, while low wind, cold weather, tighter hydro conditions, or reduced import capacity tend to push prices higher.

Geographic Scope and Development

The current implementation covers Finland (FI), Sweden (SE1-SE4), Norway (NO1-NO5), Denmark (DK1-DK2), Germany (DE), Estonia (EE), France (FR), and Poland (PL). The same weather model is used across these areas, while power, transmission, and market data come from ENTSO-e and JAO. Additionally, SYKE and NVE are utilized to capture hydology in Finland and Norway, respectively.

Limitations of the Forecast

The model only considers variables included in its training data. It does not directly account for unexpected geopolitical events, sudden market interventions, or rare structural shocks with limited historical precedent. It includes hydrology-state features for Finland and Norway, but it does not explicitly model hydro dispatch decisions or full hydro-system dynamics, and Swedish hydrology is not currently included. Treat forecasts as data-driven estimates, not future outcomes.