lect_1_match_fig

lect_1_match_fig

ID:40390243

大小:264.57 KB

页数:44页

时间:2019-08-01

lect_1_match_fig_第1页
lect_1_match_fig_第2页
lect_1_match_fig_第3页
lect_1_match_fig_第4页
lect_1_match_fig_第5页
资源描述:

《lect_1_match_fig》由会员上传分享,免费在线阅读,更多相关内容在学术论文-天天文库

1、Imbens/Wooldridge,LectureNotes1,Summer’071What’sNewinEconometricsNBER,Summer2007Lecture1,Monday,July30th,9.00-10.30amEstimationofAverageTreatmentEffectsUnderUnconfoundedness1.IntroductionInthislecturewelookatseveralmethodsforestimatingaverageeffectsofapr

2、ogram,treatment,orregime,underunconfoundedness.Thesettingisonewithabinaryprogram.Thetraditionalexampleineconomicsisthatofalabormarketprogramwheresomeindivid-ualsreceivetrainingandothersdonot,andinterestisinsomemeasureoftheeffectivenessofthetraining.Unco

3、nfoundedness,atermcoinedbyRubin(1990),referstothecasewhere(non-parametrically)adjustingfordifferencesinafixedsetofcovariatesremovesbiasesincomparisonsbetweentreatedandcontrolunits,thusallowingforacausalinterpretationofthoseadjusteddifferences.Thisisperhap

4、sthemostimportantspecialcaseforestimatingaveragetreatmenteffectsinpractice.Alternativestypicallyinvolvesstrongassumptionslink-ingunobservablestoobservablesinspecificwaysinordertoallowadjustingfortherelevantdifferencesinunobservedvariables.Anexampleofsucha

5、strategyisinstrumentalvariables,whichwillbediscussedinLecture3.AsecondexamplethatdoesnotinvolveadditionalassumptionsistheboundsapproachdevelopedbyManski(1990,2003).Underthespecificassumptionswemakeinthissetting,thepopulationaveragetreat-√menteffectcanbee

6、stimatedatthestandardparametricNratewithoutfunctionalformassumptions.Avarietyofestimators,atfirstsightquitedifferent,havebeenproposedforimplementingthis.Theestimatorsincluderegressionestimators,propensityscorebasedes-timatorsandmatchingestimators.Manyoft

7、heseareusedinpractice,althoughrarelyisthischoicemotivatedbyprincipledarguments.Inpracticethedifferencesbetweentheesti-matorsarerelativelyminorwhenappliedappropriately,althoughmatchingincombinationwithregressionisgenerallymorerobustandisprobablytherecomm

8、endedchoice.Moreim-portantthanthechoiceofestimatoraretwootherissues.Bothinvolveanalysesofthedatawithouttheoutcomevariable.First,oneshouldcarefullychecktheextentoftheoverlapImbens/Wooldridge,LectureNotes1,Summer’072incovariatedistributio

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。
相关文章
更多
相关标签