安慰劑檢驗介紹、操作及舉例.doc

《安慰劑檢驗介紹、操作及舉例.doc》由會員上傳分享，免費在線閱讀，更多相關(guān)內(nèi)容在教育資源-天天文庫。

1、安慰劑檢驗介紹（Placebotest）安慰劑是一種附加實證檢驗的思路，并不存在一個具體的特定的操作方法。一般存在兩種尋找安慰劑變量的方法。比如，在已有的實證檢驗中，發(fā)現(xiàn)自變量Xi會影響自變量Zi與因變量Yi之間存在相關(guān)關(guān)系。在其后的實證檢驗中，采用其他主體（國家，省份，公司）的Xj變量作為安慰劑變量，檢驗Xj是否影響Zi與Yi之間的相關(guān)關(guān)系。如果不存在類似于Xi的影響，即可排除Xi的安慰劑效應(yīng)，使得結(jié)果更為穩(wěn)健。另一種尋找安慰劑變量的方法。已知，Xi是虛擬變量，Xi=1，ift>T；Xi=0ift

2、DID）。在其后的實證檢驗中，將Xi`設(shè)定為Xi`=1，ift>T+n；Xi`=0ift

3、了。以下是詳細(xì)的例題，安慰劑檢驗在最后。SurvivingGraduateEconometricswithR:Difference-in-DifferencesEstimation—2of?8Thefollowingreplicationexercisecloselyfollowsthehomeworkassignment#2inECNS562.Thedataforthisexercisecanbefound?here.ThedataisabouttheexpansionoftheEarnedIncomeTaxCredit.Thisisalegislat

4、ionaimedatprovidingataxbreakforlowincomeindividuals.?Forsomebackgroundonthesubject,seeEissa,Nada,andJeffreyB.Liebman.1996.LaborSupplyResponsestotheEarnedIncomeTaxCredit.QuarterlyJournalofEconomics.?111(2):605-637.Thehomeworkquestions(abbreviated):1.Describeandsummarizedata.2.Calc

5、ulatethesamplemeansofallvariablesfor(a)singlewomenwithnochildren,(b)singlewomenwith1child,and(c)singlewomenwith2+children.3.Createanewvariablewithearningsconditionalonworking(missingfornon-employed)andcalculatethemeansofthisbygroupaswell.4.Constructavariableforthe“treatment”calle

6、dANYKIDSandavariableforaftertheexpansion(calledPOST93—shouldbe1for1994andlater).5.Createagraphwhichplotsmeanannualemploymentratesbyyear(1991-1996)forsinglewomenwithchildren(treatment)andwithoutchildren(control).6.Calculatetheunconditionaldifference-in-differenceestimatesoftheeffe

7、ctofthe1993EITCexpansiononemploymentofsinglewomen.7.Nowrunaregressiontoestimatetheconditionaldifference-in-differenceestimateoftheeffectoftheEITC.Useallwomenwithchildrenasthetreatmentgroup.8.Reestimatethismodelincludingdemographiccharacteristics.9.Addthestateunemploymentrateandal

8、lowitseffecttovarybythepresenceofchildre