the r book analysis of variance

the r book analysis of variance

ID:7294106

大小:455.52 KB

页数:39页

时间:2018-02-10

the r book analysis of variance_第1页
the r book analysis of variance_第2页
the r book analysis of variance_第3页
the r book analysis of variance_第4页
the r book analysis of variance_第5页
资源描述:

《the r book analysis of variance》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、11AnalysisofVarianceInsteadoffittingcontinuous,measuredvariablestodata(asinregression),manyexperimentsinvolveexposingexperimentalmaterialtoarangeofdiscretelevelsofoneormorecategoricalvariablesknownasfactors.Thus,afactormightbedrugtreatmentforaparticularcancer,withfivelevels

2、correspondingtoaplaceboplusfournewpharmaceuticals.Alternatively,afactormightbemineralfertilizer,wherethefourlevelsrepresentfourdifferentmixturesofnitrogen,phosphorusandpotassium.Factorsareoftenusedinexperimentaldesignstorepresentstatisticalblocks;theseareinternallyhomogen

3、eousunitsinwhicheachoftheexperimentaltreatmentsisrepeated.Blocksmaybedifferentfieldsinanagriculturaltrial,differentgenotypesinaplantphysiologyexperiment,ordifferentgrowthchambersinastudyofinsectphotoperiodism.Itisimportanttounderstandthatregressionandanalysisofvariance(ANO

4、VA)areidenticalapproachesexceptforthenatureoftheexplanatoryvariables.Forexample,itisasmallstepfromhavingthreelevelsofashadefactor(saylight,mediumandheavyshadecloths)thencarryingoutaone-wayANOVA,tomeasuringthelightintensityinthethreetreatmentsandcarryingoutaregressionwithl

5、ightintensityastheexplanatoryvariable.Asweshallseelateron,someexperimentscombineregressionandANOVAbyfittingaseriesofregressionlines,oneineachofseverallevelsofagivenfactor(thisiscalledanalysisofcovariance;seeChapter12).TheemphasisinANOVAwastraditionallyonhypothesistesting.N

6、owadays,theaimofananalysisofvarianceinRistoestimatemeansandstandarderrorsofdifferencesbetweenmeans.Comparingtwomeansbyattestinvolvedcalculatingthedifferencebetweenthetwomeans,dividingbythestandarderrorofthedifference,andthencomparingtheresultingstatisticwiththevalueofStud

7、ent’stfromtables(orbetterstill,usingqttocalculatethecriticalvalue;seep.287).Themeansaresaidtobesignificantlydifferentwhenthecalculatedvalueoftislargerthanthecriticalvalue.Forlargesamples(n>30)ausefulruleofthumbisthatatvaluegreaterthan2issignificant.InANOVA,weareconcernedwit

8、hcaseswherewewanttocomparethreeormoremeans.Forthetwo-samplecase,thettestandtheANOVAareidentical,

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。