Distributional forecasting of electricity DART spreads with a covariate-dependent mixture model

IF 13.6 2区经济学 Q1 ECONOMICS

Energy Economics Pub Date : 2025-03-03 DOI:10.1016/j.eneco.2025.108332

Anthony Forgetta , Frédéric Godin , Maciej Augustyniak

{"title":"Distributional forecasting of electricity DART spreads with a covariate-dependent mixture model","authors":"Anthony Forgetta , Frédéric Godin , Maciej Augustyniak","doi":"10.1016/j.eneco.2025.108332","DOIUrl":null,"url":null,"abstract":"<div><div>We develop a covariate-dependent mixture model to describe the behavior of electricity DART spreads, which are differentials between day-ahead and real-time prices of electricity. The model includes three regimes: a regular DART regime, a positive spike regime, and a negative spike regime. The model exhibits sufficient flexibility to allow covariates impacting both the frequency and severity of DART spread spikes, and to reproduce salient stylized facts of DART spread dynamics. The covariates considered include forecasts for load, weather, and natural gas prices. The application of our model on data from the Long Island zone of the NYISO (New York Independent System Operator) exhibits a satisfactory fit to the data. Numerical experiments reveal that including covariates in the severity component of the model is crucial, while mild additional performance is obtained with their inclusion in the frequency component. Furthermore, neural network-based quantile regression benchmarks are unable to improve performance over our mixture model.</div></div>","PeriodicalId":11665,"journal":{"name":"Energy Economics","volume":"144 ","pages":"Article 108332"},"PeriodicalIF":13.6000,"publicationDate":"2025-03-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Energy Economics","FirstCategoryId":"96","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0140988325001562","RegionNum":2,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ECONOMICS","Score":null,"Total":0}

引用次数: 0

Abstract

We develop a covariate-dependent mixture model to describe the behavior of electricity DART spreads, which are differentials between day-ahead and real-time prices of electricity. The model includes three regimes: a regular DART regime, a positive spike regime, and a negative spike regime. The model exhibits sufficient flexibility to allow covariates impacting both the frequency and severity of DART spread spikes, and to reproduce salient stylized facts of DART spread dynamics. The covariates considered include forecasts for load, weather, and natural gas prices. The application of our model on data from the Long Island zone of the NYISO (New York Independent System Operator) exhibits a satisfactory fit to the data. Numerical experiments reveal that including covariates in the severity component of the model is crucial, while mild additional performance is obtained with their inclusion in the frequency component. Furthermore, neural network-based quantile regression benchmarks are unable to improve performance over our mixture model.

查看原文本刊更多论文

求助全文

约1分钟内获得全文求助全文

来源期刊

Energy Economics ECONOMICS-

CiteScore

18.60

自引率

12.50%

发文量

524

期刊介绍： Energy Economics is a field journal that focuses on energy economics and energy finance. It covers various themes including the exploitation, conversion, and use of energy, markets for energy commodities and derivatives, regulation and taxation, forecasting, environment and climate, international trade, development, and monetary policy. The journal welcomes contributions that utilize diverse methods such as experiments, surveys, econometrics, decomposition, simulation models, equilibrium models, optimization models, and analytical models. It publishes a combination of papers employing different methods to explore a wide range of topics. The journal's replication policy encourages the submission of replication studies, wherein researchers reproduce and extend the key results of original studies while explaining any differences. Energy Economics is indexed and abstracted in several databases including Environmental Abstracts, Fuel and Energy Abstracts, Social Sciences Citation Index, GEOBASE, Social & Behavioral Sciences, Journal of Economic Literature, INSPEC, and more.