摘要：Probabilistic forecasting of electricity load curves is of fundamental importance for effective scheduling and decision making in the increasingly volatile and competitive energy markets. A critical challenge is to deal with the nonstationarity at daily, weekly and monthly levels. Working with EDF (Electricity of France), we recast the problem as a curve-to-curve regression. We propose a novel approach to construct probabilistic predictors for curves (PPC), which leads to a natural and new definition of quantiles in the context of curve-to-curve linear regression. There are three types of PPC: a predict set, a predictive band and a predictive quantile, and all of them are defined at a pre-specified nominal probability level. When applying to one day ahead forecasting for the French daily electricity load curves, PPC outperform several state-of-the-art predictive methods in terms of forecasting accuracy, coverage rate and average length of the predictive bands. The predictive quantile curves provide insightful information which is highly relevant to hedging risks in electricity supply management.

报告人简介：

Qiwei Yao is a Professor of Statistics at the London School of Economics and Political Science. He is an Honorary Member of the Royal Institution of Statistics, the American Statistical Society, and the Mathematical Statistics Society. Professor Yao is internationally renowned for his contributions to several prominent research areas, including time series analysis, space-time analysis, and financial econometrics. He serves as a co-editor of the Journal of the Royal Statistical Society (Series B) and Statistica Sinica, and has also served as an associate editor for top-tier statistical journals such as Annals of Statistics and the Journal of the American Statistical Association.