{"title":"Sources of heteroscedasticity in the spot electricity price time series","authors":"Zita Marossy","doi":"10.1109/EEM.2010.5558781","DOIUrl":null,"url":null,"abstract":"Spot electricity prices are found to be heteroscedastic in the literature. In this paper I analyze the sources of heteroscedasticity. The heteroscedasticity is measured with the autocorrelation function of the squared residuals. I will show that the heteroscedasticity effect consists of a deterministic and a stochastic part. I decompose the heteroscedasticity of power prices into three factors according to the origin of heteroscedasticity: seasonality; long memory; and GARCH behavior. I model the effects of intraweekly seasonality by using the so-called GEV filtering procedure. After removing the deterministic heteroscedasticity, the remaining heteroscedasticity can be described with a GARCH-type model. Empirical calculations show that seasonality incorporates a negative heteroscedasticity effect, i.e. the magnitude of heteroscedasticity increases after the seasonality filtering procedure.","PeriodicalId":310310,"journal":{"name":"2010 7th International Conference on the European Energy Market","volume":"29 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2010-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2010 7th International Conference on the European Energy Market","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/EEM.2010.5558781","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
Spot electricity prices are found to be heteroscedastic in the literature. In this paper I analyze the sources of heteroscedasticity. The heteroscedasticity is measured with the autocorrelation function of the squared residuals. I will show that the heteroscedasticity effect consists of a deterministic and a stochastic part. I decompose the heteroscedasticity of power prices into three factors according to the origin of heteroscedasticity: seasonality; long memory; and GARCH behavior. I model the effects of intraweekly seasonality by using the so-called GEV filtering procedure. After removing the deterministic heteroscedasticity, the remaining heteroscedasticity can be described with a GARCH-type model. Empirical calculations show that seasonality incorporates a negative heteroscedasticity effect, i.e. the magnitude of heteroscedasticity increases after the seasonality filtering procedure.