regression estimation in the presence of measurement error

regression estimation in the presence of measurement error

ID:7286129

大?。?.11 MB

頁數(shù):19頁

時間:2018-02-10

regression estimation in the presence of measurement error_第1頁
regression estimation in the presence of measurement error_第2頁
regression estimation in the presence of measurement error_第3頁
regression estimation in the presence of measurement error_第4頁
regression estimation in the presence of measurement error_第5頁
資源描述:

《regression estimation in the presence of measurement error》由會員上傳分享,免費在線閱讀,更多相關(guān)內(nèi)容在工程資料-天天文庫。

1、CHAPTER30REGRESSIONESTIMATIONINTHEPRESENCEOFMEASUREMENTERRORWayneA.FullerIowaStateUniversity30.1INTRODUCTIONRegressionisatechniqueoftenusedintheanalysisofsurveydata.Theclassicalregressionmodelisyt=x't?+qt(30.1)2wheretheqtareindependent(0,σ)randomvariablesindepend

2、entofx;foralltand;',x,isaA:-dimensionalcolumnvectorofexplanatoryvariablesandβistheunknownparametervector.Surveydatatypicallyviolatetheclassicalmodelintwoimportantways.Ifthesurveyisaclusteredsampleitisnolongerreasonabletoassumethattheqtwithinaclusterareindependent

3、.Thisfacthasbeenrecognizedforsometime.See,forexample,Frankel(1971),Fuller(1975),andFullerandHidiroglou(1978).Furthermore,computerpackageshavebeendevelopedtocomputeregressioncoefficientsandthestandarderrorsofregressioncoefficientsforcomplexsurveydata.Examplesofsuc

4、hprogramsincludeSUPERCARP,PCCARP,andSUDAAN.Thesecondimportantwayinwhichsurveydataviolatethemodelisthatonedoesnotobserve(yt,xO-Ratheroneobserves(y?XD=(*,xi)+(u>i.uDwhere(wt,uj)isthevectorofmeasurementerror.Insurveysampling,theThisresearchwaspartlysupportedbyJointS

5、tatisticalAgreementJ.S.A90-7withtheU.S.BureauoftheCensus.IthankJosephCrooeforthecomputationsofthedesignsectionandforthecomputationsoftheillustration.MeasurementErrorsinSurveys.EditedbyBiemer,Groves,Lyberg,Mathiowetz,andSudman.Copyright?1991byJohnWiley&Song,Inc.IS

6、BN:0-471-53405-6618REGRESSIONESTIMATIONWITHMEASUREMENTERRORmeasurementerrorisoftencalledresponseerrorbecauseaprimarysourceoferroristhefailureofrespondentstogivethesameanswertoarepeatedquestion.Thatsurveydataaresubjecttomeasurementerrorisuniformlyrecognizedbytheco

7、mmunityofsurveysamplers.AnearlydescriptionoftheeffectofmeasurementerroronestimatesofchangeisthatofBershad(1967).ExamplesoftheincorporationofestimatesofmeasurementerrorvariancesintoanalysesofsurveydataincludeWarren,etal.(1974),Rodgers(1989a),ChuaandFuller(1987),Po

8、terbaandSummers(1985),AbowdandZellner(1985),andnumerousarticlesinpsychologyandsociologyjournals.Despitethislistofreferences,itseemsfairtosaythatthefractionofre

當前文檔最多預覽五頁,下載文檔查看全文

此文檔下載收益歸作者所有

當前文檔最多預覽五頁,下載文檔查看全文
溫馨提示:
1. 部分包含數(shù)學公式或PPT動畫的文件,查看預覽時可能會顯示錯亂或異常,文件下載后無此問題,請放心下載。
2. 本文檔由用戶上傳,版權(quán)歸屬用戶,天天文庫負責整理代發(fā)布。如果您對本文檔版權(quán)有爭議請及時聯(lián)系客服。
3. 下載前請仔細閱讀文檔內(nèi)容,確認文檔內(nèi)容符合您的需求后進行下載,若出現(xiàn)內(nèi)容與標題不符可向本站投訴處理。
4. 下載文檔時可能由于網(wǎng)絡(luò)波動等原因無法下載或下載錯誤,付費完成后未能成功下載的用戶請聯(lián)系客服處理。