Modelling Electricity Prices in 15-Minute Intervals: A Panel Data Approach for a New Market Regime in the Electricity Pricing
Authors
Yammin, Elissa Kamil ; Pedersen, Andreas Laurids
Term
4. term
Education
Publication year
2026
Submitted on
2026-05-26
Pages
96
Abstract
This thesis examines whether panel data models can be used to model and forecast electricity prices after the shift from hourly to 15-minute settlement in Denmark’s DK1 market. Panel data econometrics combines information from many related series to uncover patterns that repeat over time (persistence and seasonality) and across series that are influenced by common factors (cross-sectional dependence). We review fixed effects and random effects (ways to handle stable differences between series), dynamic panel models (which include past prices to explain current prices), and factor-augmented models (which bring in latent common drivers). The analysis applies these methods to both hourly and 15-minute market data. The results show that 15-minute specifications improve forecasting and reveal finer market dynamics compared with hourly models. Factor-augmented models account for hidden dependence structures, but their forecasts are smoother and less reactive to abrupt shocks and extreme price spikes.
Denne afhandling undersøger, om paneldatamodeller kan bruges til at modellere og forudsige elpriser efter skiftet fra timesafregning til 15-minutters afregning i DK1-markedet i Danmark. Paneløkonometriske metoder kombinerer information fra mange beslægtede serier for at afdække mønstre, der gentager sig over tid (persistens og sæsonmønstre) og på tværs af serier, der påvirkes af fælles forhold (tværsnitafhængighed). Vi gennemgår fixed effects og random effects (måder at håndtere stabile forskelle mellem serier), dynamiske panelmodeller (som inkluderer tidligere priser for at forklare nuværende priser) og faktor-augmenterede modeller (som inddrager skjulte fælles drivkræfter). Metoderne anvendes på både time- og 15-minutters markedsdata. Resultaterne viser, at 15-minutters specifikationer forbedrer prognoserne og fanger finere markedsdynamik sammenlignet med time-modeller. Faktor-augmenterede modeller håndterer skjulte afhængighedsstrukturer, men giver glattere prognoser, der reagerer mindre på pludselige chok og ekstreme prisspidser.
[This apstract has been rewritten with the help of AI based on the project's original abstract]