Properties of Weighted Geometric means Combination Forecasting Model Based on Degree of Logarithm Grey Incidence书

《Properties of Weighted Geometric means Combination Forecasting Model Based on Degree of Logarithm Grey Incidence书》由会员上传分享，免费在线阅读，更多相关内容在学术论文-天天文库。

1、Proceedingsof2007IEEEInternationalConferenceonGreySystemsandIntelligentServices,November18-20,2007,Nanjing,ChinaPropertiesofWeightedGeometricMeansCombinationForecastingModelBasedonDegreeofLogarithmGreyIncidenceChenHuayou,DingKaimin,ZhangJunting,0Ftherearesomeotheroptimalcriteriat

2、omeasuretheAbstract—Webuildtheweightedgeometricmeanscombinationforecastingmodelbasedondegreeofprecisionofcombinationforecastingmethods.Inrecentlogarithmgrayincidence.Thisisanewkindofyears,thenewcombinationforecastingmethodsbasedonnonlinearcombinationforecastingmethod.Wecorrelativ

3、ityindexeshavebeenputforward,includingcalculateweightedcoefficientvectorsofcombinationcombinationforecastingmodelofminimizingTheilforecastingmethodsbytheoptimalcriteriaofunequalcoefficientormaximizinggrayincidence[2].maximizingthelogarithmthedegreeofgrayincidenceAnothernewmodelof

4、basicstructuralcharacteristicsofofcombinationforecastingmodel,whichisdifferentcombinationforecastingwasstudiedunderthecriterionoffromordinarycriteriaofminimizingabsoluteerrors.maximizingvectorialanglecosine[16].ThesemethodsWeputforwardsomenewconceptsofthismodel,suchwerequitediffe

5、rentfromtraditionalcombinationassuperiorcombinationforecasting,redundantdegreeforecastingmodelsinnotconsideringtheabsoluteerrorofcombinationforecastingetc.Weexhibittheofforecasting.However,thecombinationforecastingsufficientconditionsoftheexistenceofnon-inferiormodelsbasedonthede

6、greeofgrayincidencewereputandsuperiorcombinationforecasting.Wealsoproveforwardin[13],wherenumericalexampleshowedthatthethatthereexistsredundantsingleforecastingmethodrelativelygoodresultscouldalsobeobtainedbyinthecombinationforecastingifthesingleforecastingcombinationforecastingm

7、ethodsofcorrelativityindexes.methodisstrictlydominantoveranother.TheseresultsThearithmeticaveragingcombinationforecastingmodelsshowthatthecombinationforecastingmodelisonthebasisofthedegreeofgrayincidencewerebuiltineffectivetheoretically.[14]andthegoodpropertiesoflinearcombination

8、I.INTRODUCTIONforecastingmethodsbasedont