admin 管理员组文章数量: 1086019
2024年12月29日发(作者：c语言源程序)
Package‘didimputation’
October13,2022
TypePackage
TitleImputationEstimatorfromBorusyak,Jaravel,andSpiess(2021)
Version0.3.0
DescriptionEstimatesTwo-wayFixedEffectsdifference-in-differences/event-
studymodelsusingtheimputation-
basedapproachproposedbyBorusyak,Jaravel,andSpiess(2021).
EncodingUTF-8
LazyDatatrue
RoxygenNote7.2.1
LinkingToRcpp,RcppArmadillo
DependsR(>=2.10),fixest(>=0.10.0),(>=1.10.0)
ImportsMatrix,magrittr,Rcpp,broom,dplyr,glue,stringr,purrr,
tidyr
Suggestshaven,testthat(>=3.0.0)
Config/testthat/edition3
LicenseMIT+fileLICENSE
NeedsCompilationyes
AuthorKyleButts[aut,cre]()
MaintainerKyleButts<***********************>
RepositoryCRAN
Date/Publication2022-08-2520:02:33UTC
Rtopicsdocumented:
.
Index
1
2
2
3
6
2df_hom
df_hetSimulateddatawithtwotreatmentgroupsandheterogenouseffects
Description
Generatedusingthefollowingcall:did2s::gen_data(panel=c(1990,2020),g1=2000,g2=
2010,g3=0,te1=2,te2=1,te3=0,te_m1=0.05,te_m2=0.15,te_m3=0)
Usage
df_het
Format
Adataframewith31000rowsand15variables:
unitindividualinpaneldata
yeartimeinpaneldata
gtheyearthattreatmentstarts
dep_varoutcomevariable
treatT/Fvariableforwhentreatmentison
rel_=nevertreated.
rel_year_binnedyearrelativetotreatmentstart,but<=-6and>=6arebinned.
unit_feUnitFE
year_feYearFE
errorRandomerrorcomponent
teStatictreatmenteffect=te
te_dynamicDynamictreatmeteffect=te_m
stateStatethatunitisin
groupStringnameforgroup
df_homSimulateddatawithtwotreatmentgroupsandhomogenouseffects
Description
Generatedusingthefollowingcall:did2s::gen_data(panel=c(1990,2020),g1=2000,g2=
2010,g3=0,te1=2,te2=2,te3=0,te_m1=0,te_m2=0,te_m3=0)
Usage
df_hom
did_imputation
Format
Adataframewith31000rowsand15variables:
unitindividualinpaneldata
yeartimeinpaneldata
gtheyearthattreatmentstarts
dep_varoutcomevariable
treatT/Fvariableforwhentreatmentison
rel_=nevertreated.
rel_year_binnedyearrelativetotreatmentstart,but<=-6and>=6arebinned.
unit_feUnitFE
year_feYearFE
errorRandomerrorcomponent
teStatictreatmenteffect=te
te_dynamicDynamictreatmeteffect=te_m
groupStringnameforgroup
stateStatethatunitisin
weightWeightfromrunif()
3
did_imputationBorusyak,Jaravel,andSpiess(2021)Estimator
Description
Treatmenteffectestimationandpre-trendtestinginstaggeredadoptiondiff-in-diffdesignswithan
imputationapproachofBorusyak,Jaravel,andSpiess(2021)
Usage
did_imputation(
data,
yname,
gname,
tname,
idname,
first_stage=NULL,
wname=NULL,
wtr=NULL,
horizon=NULL,
pretrends=NULL,
cluster_var=NULL
)
4
Arguments
data
yname
gname
tname
idname
first_stage
wname
wtr
horizon
did_imputation
estc()syntaxformultiplelhs.
lenameforunit-specificdateoftreatment(never-treatedshould
bezeroorNA).
lenameforcalendarperiod.
lenameforuniqueunitid.
FormulaforY(0).Formulafollowingfixest::ffectsspecified
after"|".Ifnotspecified,thenjustunitandtimefixedeffectswillbeused.
usedin
estimatingY(0)andalsoaugmentstreatmenteffectweights.
Charactervectoroftreatmentweightnames(seehorizonforstandardstaticand
event-studyweights)
Integervectorofevent_lyappliesifwtrisleftasNULL.
ifspecified,weightedaverages/sumsoftreatmenteffectswillbereportedfor
eachofthesehorizonsseparately(0forthetreatmentperiod,tau1forone
periodaftertreatment,etc.).IfTRUE,ndhorizon
arenull,thenthestatictreatmenteffectiscalculated.
,allpretrendsare
used.
upplied,thenidnameisused
asdefault.
pretrends
cluster_var
Details
Theimputation-basedestimatorisamethodofcalculatingtreatmenteffectsinadifference-in-
hodestimatesamodelforY(0)usinguntreated/not-yet-treated
observationsandpredictsY(0)forthetreatedobservationshat(Y_it(0)).Thedifferencebetween
treatedandpredicteduntreatedoutcomesY_it(1)-hat(Y_it(0))servesasanestimateforthetreat-
rethenaveragedtoformaveragetreatmenteffectsfor
groupsofit.
Value
ontainingtreatmenteffectterm,estimate,standarderrorandconfidenceinterval.
Thisisintidyformat.
Examples
Loadexampledatasetwhichhastwotreatmentgroupsandhomogeneoustreatmenteffects
#LoadExampleDataset
data("df_hom",package="did2s")
StaticTWFE:
YoucanrunastaticTWFEfixedeffectmodelforasimpletreatmentindicator
did_imputation
did_imputation(data=df_hom,yname="dep_var",gname="g",
tname="year",idname="unit")
#>#Atibble:1x6
#>
#>
#>1dep_vartreat2.000.01821.972.04
EventStudy:
Oryoucanuserelative-treatmentindicatorstoestimateaneventstudyestimate
did_imputation(data=df_hom,yname="dep_var",gname="g",
tname="year",idname="unit",horizon=TRUE)
#>#Atibble:21x6
#>
#>
#>1dep_var01.970.04251.892.05
#>2dep_var12.050.04341.972.14
#>3dep_var22.030.04321.952.12
#>4dep_var31.970.04281.882.05
#>5dep_var41.970.04201.882.05
#>6dep_var52.030.04231.952.11
#>7dep_var62.040.04501.952.13
#>8dep_var72.000.04371.912.08
#>9dep_var82.020.04401.932.10
#>10dep_var91.960.04401.872.04
#>#...with11morerows
ExamplefromChengandHoekstra(2013):
Here’sanexampleusingdatafromChengandHoekstra(2013)
5
#CastleData
castle<-haven::read_dta("/scunning1975/mixtape/raw/master/")
did_imputation(data=castle,yname="c(l_homicide,l_assault)",gname="effyear",
first_stage=~0|sid+year,
tname="year",idname="sid")
#>#Atibble:2x6
#>
#>
#>1l_homicidetreat0.07980.0609-0.03950.199
#>2l_assaulttreat0.04960.0513-0.05100.150
Index
∗datasets
df_het,2
df_hom,2
df_het,2
df_hom,2
did_imputation,3
fixest::feols,4
6
版权声明：本文标题：didimputation包的说明书 内容由网友自发贡献，该文观点仅代表作者本人， 转载请联系作者并注明出处：http://www.roclinux.cn/p/1735525284a1673889.html， 本站仅提供信息存储空间服务，不拥有所有权，不承担相关法律责任。如发现本站有涉嫌抄袭侵权/违法违规的内容，一经查实，本站将立刻删除。
发表评论