商务统计全套配套课件[美]AmirD.AczelJayavelSounderpandian著谭英平等译Chap017.ppt

商务统计全套配套课件[美]AmirD.AczelJayavelSounderpandian著谭英平等译Chap017.ppt

ID:51439868

大小:350.50 KB

页数:25页

时间:2020-03-22

商务统计全套配套课件[美]AmirD.AczelJayavelSounderpandian著谭英平等译Chap017.ppt_第1页
商务统计全套配套课件[美]AmirD.AczelJayavelSounderpandian著谭英平等译Chap017.ppt_第2页
商务统计全套配套课件[美]AmirD.AczelJayavelSounderpandian著谭英平等译Chap017.ppt_第3页
商务统计全套配套课件[美]AmirD.AczelJayavelSounderpandian著谭英平等译Chap017.ppt_第4页
商务统计全套配套课件[美]AmirD.AczelJayavelSounderpandian著谭英平等译Chap017.ppt_第5页
资源描述:

《商务统计全套配套课件[美]AmirD.AczelJayavelSounderpandian著谭英平等译Chap017.ppt》由会员上传分享,免费在线阅读,更多相关内容在教育资源-天天文库

1、COMPLETE BUSINESS STATISTICSbyAMIRD.ACZEL&JAYAVELSOUNDERPANDIAN6thedition.Chapter17MultivariateAnalysisTheMultivariateNormalDistributionDiscriminantAnalysisPrincipalComponentsandFactorAnalysisUsingtheComputerMultivariateAnalysis17DescribeamultivariatenormaldistributionExplainwhenadiscriminantan

2、alysiscouldbeconductedInterprettheresultsofadiscriminantanalysisExplainwhenafactoranalysiscouldbeconductedDifferentiatebetweenprincipalcomponentsandfactorsInterpretfactoranalysisresultsLEARNINGOUTCOMES17Afterstudyingthischapter,youshouldbeableto:Ak-dimensional(vector)randomvariableX:X=(X1,X2,X3

3、...,Xk)Arealizationofak-dimensionalrandomvariableX:x=(x1,x2,x3...,xk)Ajointcumulativeprobabilitydistributionfunctionofak-dimensionalrandomvariableX:F(x1,x2,x3...,xk)=P(X1x1,X2x2,...,Xkxk)17-2TheMultivariateNormalDistributionTheMultivariateNormalDistributionf(x1,x2)x1x2PicturingtheBivariateNo

4、rmalDistributionInadiscriminantanalysis,observationsareclassifiedintotwoormoregroups,dependingonthevalueofamultivariatediscriminantfunction.X2X1Group1Group212LineLAsthefigureillustrates,itmaybeeasiertoclassifyobservationsbylookingatthemfromanotherdirection.Thegroupsappearmoreseparatedwhenview

5、edfromapointperpendiculartoLineL,ratherthanfromapointperpendiculartotheX1orX2axis.Thediscriminantfunctiongivesthedirectionthatmaximizestheseparationbetweenthegroups.17-3DiscriminantAnalysisGroup1Group2CCuttingScoreTheformoftheestimatedpredictedequation:D=b0+b1X1+b2X2+...+bkXkwherethebiarethedis

6、criminantweights.b0isaconstant.Theintersectionofthenormalmarginaldistributionsoftwogroupsgivesthecuttingscore,whichisusedtoassignobservationstogroups.ObservationswithscoreslessthanCareassignedtogroup1,andobservationswithscoresgreaterthanCareassignedtogroup2.Sincethedistributionsmayoverlap,someo

7、bservationsmaybemisclassified.Themodelmaybeevaluatedintermsofthepercentagesofobservationsassignedcorrectlyandincorrectly.TheDiscriminantFunctionDiscriminant'Repay''Assets''Debt''Famsize'.Group01Count1418SummaryofClassificationPuti

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。