统计学CH09.ppt
- 文档编号:2511724
- 上传时间:2022-10-30
- 格式:PPT
- 页数:37
- 大小:237.50KB
统计学CH09.ppt
《统计学CH09.ppt》由会员分享,可在线阅读,更多相关《统计学CH09.ppt(37页珍藏版)》请在冰豆网上搜索。
11SlideSlideChapter9ForecastingnnTimeSeriesandTimeSeriesTimeSeriesandTimeSeriesMethodsMethods(时间序列法)(时间序列法)nnComponentsofaTimeComponentsofaTimeSeriesSeries(时间序列的组成)(时间序列的组成)nnSmoothingSmoothingMethodsMethods(平滑法)(平滑法)nnTrendTrendProjectionProjection(趋势法)(趋势法)nnTrendandSeasonalTrendandSeasonalComponentsComponents(趋势和季节成分)(趋势和季节成分)nnQualitativeApproachestoQualitativeApproachestoForecastingForecasting(定性预测法)(定性预测法)22SlideSlideTimeSeriesandTimeSeriesMethodsTimeSeriesandTimeSeriesMethodsnnByreviewinghistoricaldataovertime,wecanbetterByreviewinghistoricaldataovertime,wecanbetterunderstandthepatternofpastbehaviorofavariableunderstandthepatternofpastbehaviorofavariableandbetterpredictthefuturebehavior.andbetterpredictthefuturebehavior.nnAAtimeseriestimeseriesisasetofobservationsonavariableisasetofobservationsonavariablemeasuredoversuccessivepointsintimeorovermeasuredoversuccessivepointsintimeoroversuccessiveperiodsoftime.successiveperiodsoftime.nnTheobjectiveoftimeseriesmethodsistodiscoveraTheobjectiveoftimeseriesmethodsistodiscoverapatterninthehistoricaldataandthenextrapolatethepatterninthehistoricaldataandthenextrapolatethepatternintothefuture.patternintothefuture.nnTheforecastisbasedsolelyonpastvaluesoftheTheforecastisbasedsolelyonpastvaluesofthevariableand/orpastforecasterrors.variableand/orpastforecasterrors.33SlideSlideTheComponentsofaTimeSeriesTheComponentsofaTimeSeriesnnTrendTrendComponentComponent(趋势成分)(趋势成分)ItrepresentsagradualshiftingofatimeseriestoItrepresentsagradualshiftingofatimeseriestorelativelyhigherorlowervaluesovertime.relativelyhigherorlowervaluesovertime.TrendisusuallytheresultofchangesintheTrendisusuallytheresultofchangesinthepopulation,demographics,technology,and/orpopulation,demographics,technology,and/orconsumerpreferences.consumerpreferences.nnCyclicalCyclicalComponentComponent(循环成分)(循环成分)ItrepresentsanyrecurringsequenceofpointsItrepresentsanyrecurringsequenceofpointsaboveandbelowthetrendlinelastingmorethanaboveandbelowthetrendlinelastingmorethanoneyear.oneyear.WeassumethatthiscomponentrepresentsWeassumethatthiscomponentrepresentsmultiyearcyclicalmovementsintheeconomy.multiyearcyclicalmovementsintheeconomy.44SlideSlidennSeasonalSeasonalComponentComponent(季节成分)(季节成分)Itrepresentsanyrepeatingpattern,lessthanoneItrepresentsanyrepeatingpattern,lessthanoneyearinduration,inthetimeseries.yearinduration,inthetimeseries.Thepatterndurationcanbeasshortasanhour,orThepatterndurationcanbeasshortasanhour,orevenless.evenless.nnIrregularIrregularComponentComponent(不规则成分)(不规则成分)Itisthe“catch-all”factorthataccountsfortheItisthe“catch-all”factorthataccountsforthedeviationoftheactualtimeseriesvaluefromwhatdeviationoftheactualtimeseriesvaluefromwhatwewouldexpectbasedontheothercomponents.wewouldexpectbasedontheothercomponents.Itiscausedbytheshort-term,unanticipated,andItiscausedbytheshort-term,unanticipated,andnonrecurringfactorsthataffectthetimeseries.nonrecurringfactorsthataffectthetimeseries.TheComponentsofaTimeSeriesTheComponentsofaTimeSeries55SlideSlidennMovingMovingAveragesAverages(移动平均)(移动平均)WeusetheaverageofthemostrecentWeusetheaverageofthemostrecentnndatadatavaluesinthetimeseriesastheforecastforthenextvaluesinthetimeseriesastheforecastforthenextperiod.period.Theaveragechanges,ormoves,asnewTheaveragechanges,ormoves,asnewobservationsbecomeavailable.observationsbecomeavailable.ThemovingaveragecalculationisThemovingaveragecalculationisMovingAverage=MovingAverage=(mostrecent(mostrecentnndatavalues)/datavalues)/nnUsingSmoothingMethodsinForecastingUsingSmoothingMethodsinForecasting66SlideSlidennWeightedMovingWeightedMovingAveragesAverages(加权移动平均)(加权移动平均)ThismethodinvolvesselectingweightsforeachofThismethodinvolvesselectingweightsforeachofthedatavaluesandthencomputingaweightedthedatavaluesandthencomputingaweightedmeanastheforecast.meanastheforecast.Forexample,a3-periodweightedmovingaverageForexample,a3-periodweightedmovingaveragewouldbecomputedasfollows.wouldbecomputedasfollows.FFtt+1+1=ww11(YYtt-2-2)+)+ww22(YYtt-1-1)+)+ww33(YYtt)wherethesumoftheweights(wherethesumoftheweights(wwvalues)is1.values)is1.UsingSmoothingMethodsinForecastingUsingSmoothingMethodsinForecasting77SlideSlideUsingSmoothingMethodsinForecastingUsingSmoothingMethodsinForecastingnnExponentialExponentialSmoothingSmoothing(指数平滑)(指数平滑)Itisaspecialcaseoftheweight
- 配套讲稿:
如PPT文件的首页显示word图标,表示该PPT已包含配套word讲稿。双击word图标可打开word文档。
- 特殊限制:
部分文档作品中含有的国旗、国徽等图片,仅作为作品整体效果示例展示,禁止商用。设计者仅对作品中独创性部分享有著作权。
- 关 键 词:
- 统计学 CH09
![提示](https://static.bdocx.com/images/bang_tan.gif)