reilly media] r cookbook general statistics

reilly media] r cookbook general statistics

ID:7322565

大小:668.89 KB

页数:26页

时间:2018-02-11

reilly media] r cookbook general statistics_第1页
reilly media] r cookbook general statistics_第2页
reilly media] r cookbook general statistics_第3页
reilly media] r cookbook general statistics_第4页
reilly media] r cookbook general statistics_第5页
资源描述:

《reilly media] r cookbook general statistics》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、CHAPTER9GeneralStatisticsIntroductionAnysignificantapplicationofRincludesstatisticsormodelsorgraphics.Thischapteraddressesthestatistics.Somerecipessimplydescribehowtocalculateastatistic,suchasrelativefrequency.Mostrecipesinvolvestatisticaltestsorconfidenceintervals.Thestatis

2、ticaltestsletyouchoosebetweentwocompetinghypotheses;thatparadigmisdescribednext.Confidenceintervalsreflectthelikelyrangeofapopulationparameterandarecalculatedbasedonyourdatasample.NullHypotheses,AlternativeHypotheses,andp-ValuesManyofthestatisticaltestsinthischapteruseatime-

3、testedparadigmofstatisticalinference.Intheparadigm,wehaveoneortwodatasamples.Wealsohavetwocom-petinghypotheses,eitherofwhichcouldreasonablybetrue.Onehypothesis,calledthenullhypothesis,isthatnothinghappened:themeanwasunchanged;thetreatmenthadnoeffect;yougottheexpectedanswer;t

4、hemodeldidnotimprove;andsoforth.Theotherhypothesis,calledthealternativehypothesis,isthatsomethinghappened:themeanrose;thetreatmentimprovedthepatients’health;yougotanunexpectedanswer;themodelfitbetter;andsoforth.Wewanttodeterminewhichhypothesisismorelikelyinlightofthedata:1.T

5、obegin,weassumethatthenullhypothesisistrue.2.Wecalculateateststatistic.Itcouldbesomethingsimple,suchasthemeanofthesample,oritcouldbequitecomplex.Thecriticalrequirementisthatwemustknowthestatistic’sdistribution.Wemightknowthedistributionofthesamplemean,forexample,byinvokingth

6、eCentralLimitTheorem.3.Fromthestatisticanditsdistributionwecancalculateap-value,theprobabilityofateststatisticvalueasextremeormoreextremethantheoneweobserved,whileassumingthatthenullhypothesisistrue.1954.Ifthep-valueistoosmall,wehavestrongevidenceagainstthenullhypothesis.Thi

7、siscalledrejectingthenullhypothesis.5.Ifthep-valueisnotsmallthenwehavenosuchevidence.Thisiscalledfailingtorejectthenullhypothesis.Thereisonenecessarydecisionhere:Whenisap-value“toosmall”?Inthisbook,Ifollowthecommonconventionthatwerejectthenullhypothesiswhenp<0.05andfailtorej

8、ectitwhenp>0.05.Instatisticalterminology,Ichoseasignificancelevelofα=0.05to

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。