预览加载中，请您耐心等待几秒...

基于加权双高斯分布的广义自回归条件异方差边际电价预测模型.docx

预览
1/2
2/2

在线预览结束，喜欢就下载吧，查找使用更方便

下载提示 文本预览

如果您无法下载资料，请参考说明：

1、部分资料下载需要金币，请确保您的账户上有足够的金币

2、已购买过的文档，再次下载不重复扣费

3、资料包下载后请先用软件解压，在使用对应软件打开

基于加权双高斯分布的广义自回归条件异方差边际电价预测模型 Introduction Electricitypricingplaysacrucialroleintheenergysectorandhasbecomeasignificantproblemintheelectricitymarket.Accuratepredictionoffutureelectricitypricescanhelppowersupplierstomakeappropriatedecisionsandplanforfutureoperations.Thedevelopmentofelectricitypricingpredictionmodelshasbeenthefocusofmuchresearchinrecentyears.Inthispaper,wepresentageneralizedautoregressiveconditionalheteroskedasticity(GARCH)modelbasedontheweightedbivariateGaussiandistributionforelectricitypriceprediction. LiteratureReview Severalmethodshavebeenusedforelectricitypriceprediction,includingstatisticalmodels,machinelearningapproaches,andeconometricmodels.Therehavebeenmanystudiesondifferentapproaches,suchasARIMA,SVM,andneuralnetworks,whichcarryoutaccurateelectricitypriceprediction.Furthermore,GARCH,whichisanextensionoftheARIMAmodel,hasbeenextensivelyusedtomodelthevolatilityoffinancialdata,andmorerecently,inelectricitypriceprediction. Methodology WeproposeaGARCHmodelbasedontheweightedbivariateGaussiandistributiontocapturetheconditionalheteroskedasticityofelectricityprices.Theproposedmodelallowsustoestimateboththemeanandvarianceoftheelectricitypriceseries.Itisbasedonatwo-stepprocess,wherewefirstfitabivariateGaussiandistributiontothedata,andthenapplytheweightstoestimatetheparametersoftheGARCHmodel. Yt=α0+α1*Yt-1+α2*Yt-2+β1*σt-1^2+β2*σt-2^2+εt σt^2=δ+γ1*Yt-1^2+γ2*Yt-2^2+λ1*σt-1^2+λ2*σt-2^2 whereYtistheelectricitypriceseriesattimet,σt^2istheconditionalvarianceattimet,andεtisawhitenoiseprocesswithzeromeanandunitvariance. Results TheproposedmodelwasevaluatedontheNYISOZoneAelectricitypricedatasetfromJanuary2008toDecember2018.Themodelwascomparedwithothermodels,suchasGARCHandARIMA,andtheresultsshowedthatourmodeloutperformedtheothersintermsofRootMeanSquaredError(RMSE)andMeanAbsolutePercentageError(MAPE). Conclusion Inthispaper,wehaveproposedaGARCHmodelbasedontheweightedbivariateGaussiandistributionforelectricitypriceprediction.Wehaveshownthatourmodeloutperformsothermode