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
Chapter5
DiscreteProbabilityDistributions.10.20.30.400
1
234RandomVariablesDiscreteProbabilityDistributionsExpectedValueandVarianceBinomialProbabilityDistributionPoissonProbabilityDistributionHypergeometricProbabilityDistributionArandomvariableisanumericaldescriptionoftheoutcomeofanexperiment.RandomVariablesAdiscreterandomvariablemayassumeeitherafinitenumberofvaluesoraninfinitesequenceofvalues.Acontinuousrandomvariablemayassumeanynumericalvalueinanintervalorcollectionofintervals.Letx=numberofTVssoldatthestoreinoneday, wherexcantakeon5values(0,1,2,3,4)Example:JSLAppliancesDiscreteRandomVariablewithaFiniteNumberofValuesWecancounttheTVssold,andthereisafiniteupperlimitonthenumberthatmightbesold(whichisthenumberofTVsinstock).Letx=numberofcustomersarrivinginoneday, wherexcantakeonthevalues0,1,2,...DiscreteRandomVariablewithanInfiniteSequenceofValuesWecancountthecustomersarriving,butthereisnofiniteupperlimitonthenumberthatmightarrive.Example:JSLAppliancesRandomVariablesQuestionRandomVariablexTypeFamilysizex=NumberofdependentsreportedontaxreturnDiscreteDistancefromhometostorex=DistanceinmilesfromhometothestoresiteContinuousOwndogorcatx=1ifownnopet;=2ifowndog(s)only;=3ifowncat(s)only;=4ifowndog(s)andcat(s)DiscreteTheprobabilitydistributionforarandomvariabledescribeshowprobabilitiesaredistributedoverthevaluesoftherandomvariable.Wecandescribeadiscreteprobabilitydistributionwithatable,graph,orformula.DiscreteProbabilityDistributionsTheprobabilitydistributionisdefinedbya
probabilityfunction,denotedbyf(x),whichprovidestheprobabilityforeachvalueoftherandomvariable.Therequiredconditionsforadiscreteprobabilityfunctionare:DiscreteProbabilityDistributionsf(x)>0
f(x)=1atabularrepresentationoftheprobabilitydistributionforTVsaleswasdeveloped.UsingpastdataonTVsales,… Number
UnitsSold
ofDays 0 80 1 50 2 40 3 10 4 20 200
x
f(x)0 .401 .252 .203 .054 .101.0080/200DiscreteProbabilityDistributionsExample:JSLAppliances.10.20.30.40.500
12
3
4ValuesofRandomVariablex(TVsales)ProbabilityDiscreteProbabilityDistributionsExample:JSLAppliancesGraphicalrepresentationofprobabilitydistributionDiscreteUniformProbabilityDistributionThediscreteuniformprobabilitydistributionisthesimplestexampleofadiscreteprobabilitydistributiongivenbyaformula.Thediscreteuniformprobabilityfunctionisf(x)=1/nwhere:
n=thenumberofvaluestherandom variablemayassumethevaluesoftherandomvariableareequallylikelyExpectedValueTheexpectedvalue,ormean,ofarandomvariableisameasureofitscentrallocation.Theexpectedvalueisaweightedaverageofthevaluestherandomvariablemayassume.Theweightsaretheprobabilities.Theexpectedvaluedoesnothavetobeavaluetherandomvariablecanassume.E(x)=
=
xf(x)VarianceandStandardDeviationThevariancesummarizesthevariabilityinthevaluesofarandomvariable.Thevarianceisaweightedaverageofthesquareddeviationsofarandomvariablefromitsmean.Theweightsaretheprobabilities.Var(x)=
2=
(x-
)2f(x)Thestandarddeviation,
,isdefinedasthepositivesquarerootofthevariance.expectednumberofTVssoldinaday
x
f(x)
xf(x)0 .40 .001 .25 .252 .20 .403 .05 .154 .10 .40
E(x)=1.20ExpectedValueExample:JSLAppliances01234-1.2-0.20.81.82.81.440.040.643.247.84.40.25.20.05.10.576.010.128.162.784x-
(x-
)2f(x)(x-
)2f(x)Varianceofdailysales=s2=1.660xTVssquaredStandarddeviationofdailysales=1.2884TVsVarianceExample:JSLAppliancesBinomialProbabilityDistributionFourPropertiesofaBinomialExperiment3.Theprobabilityofasuccess,denotedbyp,doesnotchangefromtrialtotrial.4.Thetrialsareindependent.2.Twooutcomes,successandfailure,arepossibleoneachtrial.1.Theexperimentconsistsofasequenceofnidenticaltrials.stationarityassumptionBinomialProbabilityDistributionOurinterestisinthenumberofsuccessesoccurringinthentrials.Weletxdenotethenumberofsuccessesoccurringinthentrials.where:
x=thenumberofsuccessesp=theprobabilityofasuccessononetrialn=thenumberoftrials
f(x)=theprobabilityofxsuccessesinn
trialsn!=n(n–1)(n–2)…..(2)(1)
BinomialProbabilityDistributionBinomialProbabilityFunctionBinomialProbabilityDistributionBinomialProbabilityFunctionProbabilityofaparticularsequenceoftrialoutcomeswithxsuccessesinntrialsNumberofexperimentaloutcomesprovidingexactlyxsuccessesinntrialsBinomialProbabilityDistributionExample:EvansElectronicsEvansElectronicsisconcernedaboutalowretentionrateforitsemployees.Inrecentyears,managementhasseenaturnoverof10%ofthehourlyemployeesannually.Choosing3hourlyemployeesatrandom,whatistheprobabilitythat1ofthemwillleavethecompanythisyear? Thus,foranyhourlyemployeechosenatrandom,managementestimatesaprobabilityof0.1thatthepersonwillnotbewiththecompanynextyear.BinomialProbabilityDistributionExample:EvansElectronicsTheprobabilityofthefirstemployeeleavingandthesecondandthirdemployeesstaying,denoted(S,F,F),isgivenby
p(1–p)(1–p)Witha.10probabilityofanemployeeleavingonanyonetrial,theprobabilityofanemployeeleavingonthefirsttrialandnotonthesecondandthirdtrialsisgivenby
(.10)(.90)(.90)=(.10)(.90)2=.081BinomialProbabilityDistributionExample:EvansElectronicsTwootherexperimentaloutcomesalsoresultinonesuccessandtwofailures.Theprobabilitiesforallthreeexperimentaloutcomesinvolvingonesuccessfollow.ExperimentalOutcome(S,F,F)(F,S,F)(F,F,S)ProbabilityofExperimentalOutcomep(1–p)(1–p)=(.1)(.9)(.9)=.081(1–p)p(1–p)=(.9)(.1)(.9)=.081(1–p)(1–p)p=(.9)(.9)(.1)=.081Total=.243BinomialProbabilityDistributionLet:p=.10,n=3,x=1Example:EvansElectronicsUsingtheprobabilityfunctionBinomialProbabilityDistribution
1stWorker2ndWorker3rdWorkerxProb.Leaves(.1)Stays(.9)32022Leaves(.1)Leaves(.1)S(.9)Stays(.9)Stays(.9)S(.9)S(.9)S(.9)L(.1)L(.1)L(.1)L(.1).0010.0090.0090.7290.009011.0810.0810.08101Example:EvansElectronicsUsingatreediagramBinomialProbabilitiesandCumulativeProbabilitiesWithmoderncalculatorsandthecapabilityofstatisticalsoftwarepackages,suchtablesarealmostunnecessary.Thesetablescanbefoundinsomestatisticstextbooks.Statisticianshavedevelopedtablesthatgiveprobabilitiesandcumulativeprobabilitiesforabinomialrandomvariable.BinomialProbabilityDistributionE(x)=
=npVar(x)=
2=np(1-
p)ExpectedValueVarianceStandardDeviationBinomialProbabilityDistributionE(x)=np
=3(.1)=.3employeesoutof3Var(x)=np(1–p)=3(.1)(.9)=.27ExpectedValueVarianceStandardDeviationExample:EvansElectronicsAPoissondistributedrandomvariableisoftenusefulinestimatingthenumberofoccurrencesoveraspecifiedintervaloftimeorspaceItisadiscreterandomvariablethatmayassumeaninfinitesequenceofvalues(x=0,1,2,...).PoissonProbabilityDistributionExamplesofaPoissondistributedrandomvariable:thenumberofknotholesin14linearfeetofpineboardthenumberofvehiclesarrivingatatollboothinonehourPoissonProbabilityDistributionBellLabsusedthePoissondistributiontomodelthearrivalofphonecalls.PoissonProbabilityDistributionTwoPropertiesofaPoissonExperimentTheoccurrenceornonoccurrenceinanyintervalisindependentoftheoccurrenceornonoccurrenceinanyotherinterval.Theprobabilityofanoccurrenceisthesameforanytwointervalsofequallength.PoissonProbabilityFunctionPoissonProbabilityDistributionwhere:
x=thenumberofoccurrencesinanintervalf(x)=theprobabilityofxoccurrencesinaninterval
=meannumberofoccurrencesinaninterval
e
=2.71828x!=x(x–1)(x–2)...(2)(1)PoissonProbabilityDistributionPoissonProbabilityFunctionInpracticalapplications,xwilleventuallybecomelargeenoughsothatf(x)isapproximatelyzeroandtheprobabilityofanylargervaluesofxbecomesnegligible.Sincethereisnostatedupperlimitforthenumberofoccurrences,theprobabilityfunctionf(x)isapplicableforvaluesx=0,1,2,…withoutlimit.PoissonProbabilityDistributionExample:MercyHospitalPatientsarriveattheemergencyroomofMercyHospitalattheaveragerateof6perhouronweekendevenings.Whatistheprobabilityof4arrivalsin30minutesonaweekendevening?PoissonProbabilityDistribution
=6/hour=3/half-hour,x=4Example:MercyHospitalUsingtheprobabilityfunctionPoissonProbabilityDistributionPoissonProbabilities0.000.050.100.150.200.25012345678910NumberofArrivalsin30MinutesProbabilityactually,thesequencecontinues:11,12,…Example:MercyHospitalPoissonProbabilityDistributionApropertyofthePoissondistributionisthatthemeanandvarianceareequal.
m=s2PoissonProbabilityDistributionVarianceforNumberofArrivals During30-MinutePeriods m=s
2=3Example:MercyHospitalHypergeometricProbabilityDistributionThehypergeometricdistributioniscloselyrelatedtothebinomialdistribution.However,forthehypergeometricdistribution:thetrialsarenotindependent,andtheprobabilityofsuccesschangesfromtrialtotrial.HypergeometricProbabilityFunctionHypergeometricProbabilityDistributionwhere:
x
=numberofsuccesses
n
=numberoftrialsf(x)=probabilityofxsuccessesinntrials
N=numberofelementsinthepopulation
r=numberofelementsinthepopulation labeledsuccessHypergeometricProbabilityFunctionHypergeometricProbabilityDistributionfor0<
x
<
rnumberofwaysxsuccessescanbeselectedfromatotalofrsuccessesinthepopulationnumberofwaysn–xfailurescanbeselectedfromatotalofN–rfailuresinthepopulationnumberofwaysnelementscanbeselectedfromapopulationofsizeNHypergeometricProbabilityDistributionHypergeometricProbabilityFunctionIfthesetwoconditionsdonotholdforavalueofx,thecorrespondingf(x)equals0.However,onlyvaluesofxwhere:1)x
<
rand2)n–x
<
N–rarevalid.Theprobabilityfunctionf(x)onthepreviousslideisusuallyapplicableforvaluesofx=0,1,2,…n.HypergeometricProbabilityDistributionBobNevereadyhasremovedtwodeadbatteriesfromaflashlightandinadvertentlymingledthemwiththetwogoodbatteriesheintendedasreplacem
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
- 二零一四年度高端装备制造技术许可合同
- 2024年度停车场车位转让合同
- 2024版智能手表区域销售代理协议
- 2024年度材料员岗位聘请协议
- 2024年度智能办公系统集成与实施合同
- 二零二四年度渔业养殖专业培训合同
- 2024年度废旧物资回收与处理服务合同
- 2024年度烘焙店租赁合同书
- 二零二四年度劳动合同服务条款和工资福利
- 二零二四年度大型商场物业管理合同
- 水平仪的读数方法
- 4-5《地球家园的化学变化》精编课件
- 小学英语单词完整版
- 重庆市社会公共安全视频图像信息系统技术规范
- 2022年03月浙江嘉兴南湖区教育研究培训中心选聘研训员笔试参考题库答案解析版
- 有经营才有结果必须开个人酒会月日
- 《新疆维吾尔自治区去极端化条例》(全文)及教案
- GB/T 32722-2016土壤质量土壤样品长期和短期保存指南
- GB 5606.5-2005卷烟第5部分:主流烟气
- 幼儿园食堂安全知识培训测试题附答案
- 《第一单元 行进之歌-欣赏-☆中国人民解放军进行曲课件》初中音乐人音版七年级下册221
评论
0/150
提交评论