New framework forecasts non-stationary electricity prices using Langevin and NODE

Researchers have developed a new forecasting framework that combines an N-dimensional Langevin equation with a neural-ordinary differential equation (NODE). This approach is designed to model and predict non-stationary electricity price time series, which are common in electricity markets but often overlooked by existing stationary techniques. The Langevin equation captures fine-grained details in stationary conditions, while the NODE learns and predicts the differences, thereby reconstructing the non-stationary components that the Langevin equation cannot. The framework was tested using the Spanish electricity day-ahead market, demonstrating its effectiveness in handling both stationary and non-stationary price behaviors. AI

IMPACT This novel approach could improve the accuracy of financial forecasting in volatile markets.

RANK_REASON The cluster contains an academic paper detailing a new forecasting methodology. [lever_c_demoted from research: ic=1 ai=0.7]

Read on arXiv cs.LG →

• Antonio Malpica-Morales
• arXiv
• Langevin equation
• N-dimensional Langevin equation
• Neural Ordinary Differential Equations for Grey-Box Modelling of Lithium-Ion Batteries on the Basis of an Equivalent Circuit Model
• Spanish electricity day-ahead market

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →