Probabilistic Forecasting

Probabilistic forecasting aims to predict not only the most likely future value of a time series (e.g., electricity price, wind power output) but also its associated uncertainty, typically represented as a probability distribution. Current research emphasizes improving the accuracy and reliability of these probability distributions, focusing on techniques like quantile regression, conformal prediction, and generative models, often within ensemble frameworks or enhanced by deep learning architectures. These advancements are crucial for robust decision-making in various sectors, particularly those involving risk management and resource allocation under uncertainty, such as energy markets and financial planning. The development of methods that handle missing data and maintain coherence across different aggregation levels of forecasts is also a significant area of ongoing investigation.

Papers

March 29, 2024

Conformal Prediction for Stochastic Decision-Making of PV Power in Electricity Markets
Yvet Renkema, Nico Brinkel, Tarek Alskaif
Conformal Prediction Electricity Market Photovoltaic System Day Ahead Conformal Predictive Probabilistic Forecasting Stochastic Decision

March 6, 2024

Tackling Missing Values in Probabilistic Wind Power Forecasting: A Generative Approach
Honglin Wen, Pierre Pinson, Jie Gu, Zhijian Jin
Generative Approach Parameter Estimation Policy Value Wind Power Forecasting Task Probabilistic Forecasting

November 13, 2023

A probabilistic forecast methodology for volatile electricity prices in the Australian National Electricity Market
Cameron Cornell, Nam Trong Dinh, S. Ali Pourmousavi
Probabilistic Forecast Electricity Market Probabilistic Forecasting

October 9, 2023

Deep Optimal Timing Strategies for Time Series
Chen Pan, Fan Zhou, Xuanwei Hu, Xinxin Zhu, Wenxin Ning, Zi Zhuang, Siqiao Xue, James Zhang, Yunhua Hu
Time Series Probabilistic Time Series Forecasting Optimal Timing Probabilistic Forecasting Execution Time

June 5, 2023

Non-parametric Probabilistic Time Series Forecasting via Innovations Representation
Xinyi Wang, Meijen Lee, Qing Zhao, Lang Tong
Probabilistic Time Series Forecasting Semi Parametric Novel Representation Probabilistic Forecasting

October 17, 2022

Probabilistic Forecasting Methods for System-Level Electricity Load Forecasting
Philipp Giese
Long Term Key Factor Load Forecasting Energy Policy Research Probabilistic Load Probabilistic Forecasting

February 17, 2022

Ensemble Conformalized Quantile Regression for Probabilistic Time Series Forecasting
Vilde Jensen, Filippo Maria Bianchi, Stian Norman Anfinsen
Time Series Probabilistic Time Series Forecasting Conformal Predictor Conformalized Quantile Regression Probabilistic Forecasting

November 5, 2021

Coherent Probabilistic Aggregate Queries on Long-horizon Forecasts
Prathamesh Deshpande, Sunita Sarawagi
Long Horizon Time Series Forecasting Method Level Aggregation Probabilistic Forecasting

Probabilistic Forecasting

Papers

Conformal Prediction for Stochastic Decision-Making of PV Power in Electricity Markets

Tackling Missing Values in Probabilistic Wind Power Forecasting: A Generative Approach

A probabilistic forecast methodology for volatile electricity prices in the Australian National Electricity Market

Deep Optimal Timing Strategies for Time Series

Non-parametric Probabilistic Time Series Forecasting via Innovations Representation

Probabilistic Forecasting Methods for System-Level Electricity Load Forecasting

Ensemble Conformalized Quantile Regression for Probabilistic Time Series Forecasting

Coherent Probabilistic Aggregate Queries on Long-horizon Forecasts