商务统计相关知识(英文版).ppt

《商务统计相关知识(英文版).ppt》由会员上传分享，免费在线阅读，更多相关内容在工程资料-天天文库。

1、本资料来源Chapter8IntervalEstimationPopulationMean:sKnownPopulationMean:sUnknownDeterminingtheSampleSizePopulationProportionApointestimatorcannotbeexpectedtoprovidetheexactvalueofthepopulationparameter.Anintervalestimatecanbecomputedbyaddingandsubtractingamarginoferro

2、rtothepointestimate.PointEstimate+/-MarginofErrorThepurposeofanintervalestimateistoprovideinformationabouthowclosethepointestimateistothevalueoftheparameter.MarginofErrorandtheIntervalEstimateThegeneralformofanintervalestimateofapopulationmeanisMarginofErrorandthe

3、IntervalEstimatetofindouttwopositivenumbersδandα,thelatterlyingbetween0and1,suchthattheprobabilitythattherandomIntervalcontainsthetrueparameteris1-αTherandomintervalisConfidenceintervalSuchaninterval,ifitexists,isknownasaconfidenceinterval;1−αisknownastheconfidenc

4、ecoefficient;andα(0<α<1)isknownasthelevelofsignificance.TheendpointsoftheconfidenceintervalareknownastheconfidencelimitsLowerconfidencelimitsandUpperconfidencelimitsIntervalEstimationofaPopulationMean:sKnownInordertodevelopanintervalestimateofapopulationmean,thema

5、rginoferrormustbecomputedusingeither:thepopulationstandarddeviations,orthesamplestandarddeviationssisrarelyknownexactly,butoftenagoodestimatecanbeobtainedbasedonhistoricaldataorotherinformation.Werefertosuchcasesasthesknowncase.IntervalEstimateofmIntervalEstimateo

6、faPopulationMean:sKnownwhere:isthesamplemean1-istheconfidencecoefficientz/2isthezvalueprovidinganareaof/2intheuppertailofthestandardnormalprobabilitydistributionsisthepopulationstandarddeviationnisthesamplesize(7.1)Itisveryimportanttoknowthefollowingaspectsofin

7、tervalestimation:Equation(7.1)doesnotsaythattheprobabilityoftruemlyingbetweenthegivenlimitsis1−α.Sincem,althoughanunknown,isassumedtobesomefixednumber,eitheritliesintheintervaloritdoesnot.What(7.1)statesisthat,forthemethoddescribedinthischapter,theprobabilityofcon

8、structinganintervalthatcontainsmis1−α.ConfidencetothemethoditselfnottheintervalItisveryimportanttoknowthefollowingaspectsofintervalestimation:Theinterva