版权说明:本文档由用户提供并上传,收益归属内容提供方,若内容存在侵权,请进行举报或认领
文档简介
Statisticsfor
BusinessandEconomics(14e)
MetricVersionAnderson,Sweeney,Williams,Camm,Cochran,Fry,Ohlmann©2020CengageLearning©2020Cengage.Maynotbescanned,copiedorduplicated,orpostedtoapubliclyaccessiblewebsite,inwholeorinpart,exceptforuseaspermittedinalicensedistributedwithacertainproductorserviceorotherwiseonapassword-protectedwebsiteorschool-approvedlearningmanagementsystemforclassroomuse.1Chapter7-SamplingandSamplingDistributions
2Introduction(1of2)Anelementistheentityonwhichdataarecollected.Apopulationisacollectionofalltheelementsofinterest.Asampleisasubsetofthepopulation.Thesampledpopulationisthepopulationfromwhichthesampleisdrawn.Aframeisalistoftheelementsthatthesamplewillbeselectedfrom.3Introduction(2of2)Thereasonweselectasampleistocollectdatatoansweraresearchquestionaboutapopulation.Thesampleresultsprovideonlyestimatesofthevaluesofthepopulationcharacteristics.Thereasonissimplythatthesamplecontainsonlyaportionofthepopulation.Withpropersamplingmethods,thesampleresultscanprovide“good”estimatesofthepopulationcharacteristics.4SamplingfromaFinitePopulation(1of2)Finitepopulationsareoftendefinedbylistssuchas:OrganizationmembershiprosterCreditcardaccountnumbersInventoryproductnumbersAsimplerandomsampleofsizenfromafinitepopulationofsizeNisasampleselectedsuchthateachpossiblesampleofsizenhasthesameprobabilityofbeingselected.Replacingeachsampledelementbeforeselectingsubsequentelementsiscalledsamplingwithreplacement.Anelementcanappearinthesamplemorethanonce.Samplingwithoutreplacementistheprocedureusedmostoften.Inlargesamplingprojects,computer-generatedrandomnumbersareoftenusedtoautomatethesampleselectionprocess.5SamplingfromaFinitePopulation(2of2)St.Andrew’sCollegereceived900applicationsforadmissionintheupcomingyearfromprospectivestudents.Theapplicantswerenumbered,from1to900,astheirapplicationsarrived.TheDirectorofAdmissionswouldliketoselectasimplerandomsampleof30applicants.Step1: Assignarandomnumbertoeachofthe900applicants. TherandomnumbersgeneratedbyExcel’sRANDfunctionfollowauniformprobabilitydistributionbetween0and1.Step2: Selectthe30applicantscorrespondingtothe30smallestrandomnumbers.6SamplingfromanInfinitePopulation(1of3)Sometimeswewanttoselectasample,butfindthatitisnotpossibletoobtainalistofallelementsinthepopulation.Asaresult,wecannotconstructaframeforthepopulation.Hencewecannotusetherandomnumberselectionprocedure.Mostoftenthissituationoccursinthecaseofinfinitepopulation.7SamplingfromanInfinitePopulation(2of3)Populationsareoftengeneratedbyanongoingprocesswherethereisnoupperlimitonthenumberofunitsthatcanbegenerated.Someexamplesofon-goingprocesseswithinfinitepopulationsare:partsbeingmanufacturedonaproductionlinetransactionsoccurringatabanktelephonecallsarrivingatatechnicalhelpdeskcustomersenteringastore8SamplingfromanInfinitePopulation(3of3)Inthecaseofaninfinitepopulation,wemustselectarandomsampleinordertomakevalidstatisticalinferencesaboutthepopulationfromwhichthesampleistaken.Arandomsamplefromaninfinitepopulationisasampleselectedsuchthatthefollowingconditionsaresatisfied.Eachelementselectedcomesfromthepopulationofinterest.Eachelementisselectedindependently.9PointEstimation(1of4)Pointestimationisaformofstatisticalinference.Inpointestimationweusethedatafromthesampletocomputeavalueofasamplestatisticthatservesasanestimateofapopulationparameter.10PointEstimation(2of4)St.Andrew’sCollegereceived900applicationsfromprospectivestudents.Theapplicationformcontainsavarietyofinformationincludingtheindividual’sScholasticAptitudeTest(SAT)scoreandwhetherornottheindividualdesireson-campushousing.Atameetinginafewhours,theDirectorofAdmissionswouldliketoannouncetheaverageSATscoreandtheproportionofapplicantsthatwanttoliveoncampus,forthepopulationof900applicants.Thedataontheapplicantshavenotyetbeenenteredinthecollege’sdatabase.SotheDirectordecidestoestimatethevaluesofthepopulationparametersofinterestbasedonsamplestatistics.Asampleof30applicantsisselectedusingcomputer-generatedrandomnumbers.11PointEstimation(3of4)
Note:Differentrandomnumberswouldhaveidentifiedadifferentsamplewhichwouldhaveresultedindifferentpointestimates.12PointEstimation(4of4)Onceallthedataforthe900applicantswereenteredinthedatabaseofthecollege,thevaluesofthepopulationparametersofinterestwerecalculated.PopulationMeanSATScore:PopulationStandardDeviationforSATScore:PopulationproportionwantingOn-CampusHousing:13SummaryofPointEstimatesObtainedfromaSimpleRandomSamplePopulationParameterParameterValuePointEstimatorPointEstimateμ=PopulationmeanSATscore16971684σ=Populationstd.deviationforSATscore87.4s=Samplestd.deviationforSATscore85.2p=Populationproportionwantingcampushousing0.720.6714PracticalAdviceThetargetpopulationisthepopulationwewanttomakeinferencesabout.Thesampledpopulationisthepopulationfromwhichthesampleisactuallytaken.Wheneverasampleisusedtomakeinferencesaboutapopulation,weshouldmakesurethatthetargetedpopulationandthesampledpopulationareincloseagreement.15
ProcessofStatisticalInference16
Whentheexpectedvalueofthepointestimatorequalsthepopulationparameter,wesaythepointestimatorisunbiased.17
18
Afinitepopulationistreatedasbeinginfiniteifisthefinitepopulationcorrectionfactor.isreferredtoasthestandarderrorofthemean.19
20CentralLimitTheorem
21
Example:St.Andrew’sCollege22
23
Example:St.Andrew’sCollegeStep1:Calculatethez-valueattheupperendpointoftheinterval.Step2:Findtheareaunderthecurvetotheleftoftheupperendpoint.24
Example:St.Andrew’sCollegeCumulativeProbabilitiesfortheStandardNormalDistributionz.00.01.02.03.04.......5.6915.6950.6985.7019.7054.6.7257.7291.7324.737.7389.7.7580.7611.7642.7673.7704.8.7881.7910.7939.7967.7995.9.8159.8186.8212.8238.826425
Example:St.Andrew’sCollegeStep3:Calculatethez-valueatthelowerendpointoftheinterval.Step4:Findtheareaunderthecurvetotheleftofthelowerendpoint.26
Example:St.Andrew’sCollegeStep5:Calculatetheareaunderthecurvebetweenthelowerandupperendpointsoftheinterval.TheprobabilitythattheestimateofpopulationmeanSATscorewillbebetween1687and1707is:27
Example:St.Andrew’sCollegeSupposeweselectasimplerandomsampleof100applicantsinsteadofthe30originallyconsidered.
28
Example:St.Andrew’sCollege29
Example:St.Andrew’sCollege30
Example:St.Andrew’sCollege31
MakingInferencesaboutaPopulationProportion32
where:p=thepopulationproportion33
34
35
Example:St.Andrew’sCollegeRecallthat72%oftheprospectivestudentsapplyingtoSt.Andrew’sCollegedesireon-campushousing.Whatistheprobabilitythatasimplerandomsampleof30applicantswillprovideanestimateofthepopulationproportionofapplicantdesiringon-campushousingthatiswithinplusorminus.05oftheactualpopulationproportion?36
Forourexample,withn=30andp=.72,thenormaldistributionisanacceptableapproximationbecause37
Example:St.Andrew’sCollegeStep1:Calculatethez-valueattheupperendpointoftheinterval.Step2:Findtheareaunderthecurvetotheleftoftheupperendpoint.38
Example:St.Andrew’sCollegeCumulativeProbabilitiesfortheStandardNormalDistributionz.00.01.02.03.04.......5.6915.6950.6985.7019.7054.6.7257.7291.7324.7387.7389.7.7580.7611.7642.7673.7704.8.7881.7910.7939.7967.7995.9.8159.8186.8212.8238.8264......39
Example:St.Andrew’sCollegeStep3:Calculatethez-valueatthelowerendpointoftheinterval.Step4:Findtheareaunderthecurvetotheleftofthelowerendpoint.40
41OtherSamplingMethodsStratifiedRandomSamplingClusterSamplingSystematicSamplingConvenienceSamplingJudgmentSampling42StratifiedRandomSamplingThepopulationisfirstdividedintogroupsofelementscalledstrata.Eachelementinthepopulationbelongstooneandonlyonestratum.Bestresultsareobtainedwhentheelementswithineachstratumareasmuchalikeaspossible(i.e.,ahomogeneousgroup).43StratifiedRandomSampling,Part2Asimplerandomsampleistakenfromeachstratum.Formulasareavailableforcombiningthestratumsampleresultsintoonepopulationparameterestimate.Advantage:Ifstrataarehomogeneous,thismethodprovidesresultsthatareas“precise”assimplerandomsamplingbutwithasmallertotalsamplesize.Example:Thebasisforformingthestratamightbedepartment,location,age,industrytype,andsoon.44ClusterSampling(1of2)Thepopulationisfirstdividedintoseparategroupsofelementscalledclusters.Ideally,eachclusterisarepresentativesmall-scaleversionofthepopulation(i.e.,heterogeneousgroup).Asimplerandomsampleoftheclustersisthentaken.Allelementswithineachsampled(chosen)clusterformthesample.45ClusterSampling(2of2)Example:Aprimaryapplicationisareasampling,whereclustersarecityblocksorotherwell-definedareas.Advantage:Thecloseproximityofelementscanbecosteffective(i.e.,manysampleobservationscanbeobtainedinashorttime).Disadvantage:Thismethodgenerallyrequiresalargertotalsamplesizethansimpleorstratifiedrandomsampling.46SystematicSampling(1of2)IfasamplesizeofnisdesiredfromapopulationcontainingNelements,wemightsampleoneelementforeveryN/nelementsinthepopulation.WerandomlyselectoneofthefirstN/nelementsfromthepopulationlist.WethenselecteveryN/nthelementthatfollowsinthepopulationlist.47SystematicSampling(2of2)Thismethodhasthepropertiesofasimplerandomsample,especiallyifthelistofthepopulationelementsisarandomordering.Advantage:Thesampleusuallywillbeeasiertoidentifythanitwouldbeifsimplerandomsamplingwereused.Example:Selectingevery100thlistinginatelephonebookafterthefirstrandomlyselectedlisting.48ConvenienceSamplingItisanonprobabilitysamplingtechnique.Itemsareincludedinthesamplewithoutknownprobabilitiesofbeingselected.Thesample
温馨提示
- 1. 本站所有资源如无特殊说明,都需要本地电脑安装OFFICE2007和PDF阅读器。图纸软件为CAD,CAXA,PROE,UG,SolidWorks等.压缩文件请下载最新的WinRAR软件解压。
- 2. 本站的文档不包含任何第三方提供的附件图纸等,如果需要附件,请联系上传者。文件的所有权益归上传用户所有。
- 3. 本站RAR压缩包中若带图纸,网页内容里面会有图纸预览,若没有图纸预览就没有图纸。
- 4. 未经权益所有人同意不得将文件中的内容挪作商业或盈利用途。
- 5. 人人文库网仅提供信息存储空间,仅对用户上传内容的表现方式做保护处理,对用户上传分享的文档内容本身不做任何修改或编辑,并不能对任何下载内容负责。
- 6. 下载文件中如有侵权或不适当内容,请与我们联系,我们立即纠正。
- 7. 本站不保证下载资源的准确性、安全性和完整性, 同时也不承担用户因使用这些下载资源对自己和他人造成任何形式的伤害或损失。
最新文档
- 球墨铸铁管安装施工方案及工艺方法
- 核医学科甲状腺功能异常核医学检查指南
- 既有涵洞加固施工方案
- 2026年BBO生物竞赛真题及答案解析
- 高处坠落事故处置方案
- 未来五年钢带焊管市场需求变化趋势与商业创新机遇分析研究报告
- 未来五年沿海旅客运输市场需求变化趋势与商业创新机遇分析研究报告
- 2026云南楚雄州永仁县发展和改革局政府购买服务人员招聘5人备考题库及参考答案详解(完整版)
- 2026江苏淮安市淮阴师范学院部分教师岗招聘4人备考题库及参考答案详解(综合题)
- 2026春季福建泉州市晋江市第五实验小学语文自聘教师招聘2人备考题库附参考答案详解(典型题)
- 工程质量验收规范练习题及答案
- 2026年体育场馆物业赛事活动保障方案
- 2025年北京市各区高三语文一模作文范文汇编(议论文部分)
- 发电公司现货交易奖惩制度
- 2026年机关事务管理局遴选笔试试题及参考答案
- 2022年全国森林、草原、湿地调查监测技术规程-附录
- 安徽省2024年中考化学真题(含答案)
- 第十五届全国交通运输行业“极智杯”公路收费及监控员职业技能大赛考试题库-上(单选题部分)
- 基础护理学-第十一章-排泄试题及答案
- 船舶与海上技术 液化天然气燃料船舶加注规范
- 物控部绩效考核办法培训课件
评论
0/150
提交评论