forked from zhaoxiliang/causalinference
-
Notifications
You must be signed in to change notification settings - Fork 5
/
Copy pathRcode_基本有用的计量经济学.R
528 lines (454 loc) · 24.3 KB
/
Rcode_基本有用的计量经济学.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
#=========================================================================#
# Author Information : #
# Name: 李建成 Li Jian-cheng #
# Unit: Sun Yat-Sen University #
# Email: [email protected] #
# Address: No. 135, Xingang Xi Road, Guangzhou, 510275, P. R. China #
# STATEMENT: #
# This code is just preparing for study and communication. #
# It is not allowed for any commercial purposes. #
# There are any quesitons or problems, please cantact to author by e-mail. #
# The code is not perfect, it is welcome to argue. #
#=========================================================================#
setwd("E:\\EDUCATION\\STATA\\基本有用的计量经济学\\programs")
library(readstata13) # package for reading the data created by stata13~15
library(plyr) # package for data
#-----------------------------------------------#
#-------Chapter5: Linear Model Regression-------#
#-----------------------------------------------#
chip2002 <- read.dta13("ch5\\chip2002.dta")
summary(chip2002) # Statistical description; result1 in P73
chip2002$city <- as.factor(chip2002$city)
dummycity <- model.matrix(~city, data=chip2002) # creat the dummy variables of city
chip2002$industry <- as.factor(chip2002$industry)
dummyindustry <- model.matrix(~industry, data=chip2002) # creat the dummy variables of industry
chip2002 <- cbind(chip2002,dummycity,dummyindustry) # combine the chip2002 and the dummy variables
linear_fit1 <- lm(lwage~educ+exper+expersq,data=chip2002) # 方程1：lwage=c+educ+exper+expersq+u1
summary(linear_fit1) # result1 in P74
# Call:
# lm(formula = lwage ~ educ + exper + expersq, data = chip2002)
#
# Residuals:
# Min 1Q Median 3Q Max
# -7.2345 -0.3229 0.0649 0.4098 2.7205
#
# Coefficients:
# Estimate Std. Error t value Pr(>|t|)
# (Intercept) 7.544e+00 4.259e-02 177.12 < 2e-16 ***
# educ 8.734e-02 2.505e-03 34.86 < 2e-16 ***
# exper 3.728e-02 2.940e-03 12.68 < 2e-16 ***
# expersq -3.619e-04 7.267e-05 -4.98 6.46e-07 ***
# ---
# Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
#
# Residual standard error: 0.6926 on 9256 degrees of freedom
# Multiple R-squared: 0.1584, Adjusted R-squared: 0.1581
# F-statistic: 580.7 on 3 and 9256 DF, p-value: < 2.2e-16
#FWL定理--A simple proof
test_fit1 <- lm(lwage~exper+expersq,data=chip2002) # 方程2：lwage=c+exper+expersq+u2
chip2002$fity <- test_fit1$residuals
test_fit2 <- lm(educ~exper+expersq,data=chip2002) # 方程3：educ=c+exper+expersq+u3
chip2002$fitx <- test_fit2$residuals
test_fitFWL1 <- lm(fity~fitx,data=chip2002) # 方程FWL1：u2=c+u3+u'
summary(test_fitFWL1)
# Call:
# lm(formula = fity ~ fitx, data = chip2002)
#
# Residuals:
# Min 1Q Median 3Q Max
# -7.2345 -0.3229 0.0649 0.4098 2.7205
#
# Coefficients:
# Estimate Std. Error t value Pr(>|t|)
# (Intercept) -1.881e-16 7.196e-03 0.00 1
# fitx 8.734e-02 2.505e-03 34.87 <2e-16 ***
# ---
# Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
#
# Residual standard error: 0.6925 on 9258 degrees of freedom
# Multiple R-squared: 0.1161, Adjusted R-squared: 0.116
# F-statistic: 1216 on 1 and 9258 DF, p-value: < 2.2e-16
test_fitFWL2 <- lm(lwage~fitx,data=chip2002) # 方程FWL2：lwage=u3+u''
summary(test_fitFWL2)
# Call:
# lm(formula = lwage ~ 0 + fitx, data = chip2002)
#
# Residuals:
# Min 1Q Median 3Q Max
# 1.914 8.788 9.198 9.559 11.648
#
# Coefficients:
# Estimate Std. Error t value Pr(>|t|)
# fitx 0.08734 0.03312 2.637 0.00838 **
# ---
# Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
#
# Residual standard error: 9.156 on 9259 degrees of freedom
# Multiple R-squared: 0.0007504, Adjusted R-squared: 0.0006425
# F-statistic: 6.954 on 1 and 9259 DF, p-value: 0.008379
#plot x & y
plot(chip2002$educ,chip2002$lwage)
plot(chip2002$fitx,chip2002$fity)
city <- names(chip2002)[16:76]
industry <- names(chip2002)[78:88]
linear_fit2 <- lm(lwage~educ+exper+expersq,data=chip2002)
linear_fit3 <- lm(lwage~educ+exper+expersq+male+city,data=chip2002)
linear_fit4 <- lm(lwage~educ+exper+expersq+male+city+industry,data=chip2002) #result in P75
#--------------------------------------------------------------------------#
#-------Chapter6: Matching Method and Estimation of Treatment Effect-------#
#--------------------------------------------------------------------------#
library(Matching)
library(plm)
nsw_dw <- read.dta13("ch6\\nsw_dw.dta")
cps_controls <- read.dta13("ch6\\cps_controls.dta")
new_cps <- rbind(nsw_dw[nsw_dw$treat==1,],cps_controls)
##propensity score matching using nsw_dw
#ATT
logit_m1 <- glm(treat~age+education+black+hispanic+married+nodegree+re74+re75,data=nsw_dw,family = binomial(link = "logit"))
match_ATT1 <- Match(Y=nsw_dw$re78,Tr=nsw_dw$treat,X=logit_m1$fitted,estimand="ATT",M=1,replace=T)
match_ATT_adj1 <- Match(Y=nsw_dw$re78,Tr=nsw_dw$treat,Z=nsw_dw$re75,X=logit_m$fitted,estimand="ATT",M=1,BiasAdjust = T,replace=T)
summary(match_ATT1 )
summary(match_ATT_adj1 )
# kernel density graph
library(sm)
#before match
densitypscore1 <- data.frame(pscore=logit_m1$fitted,treat=nsw_dw$treat)
sm.density.compare(densitypscore1$pscore,group = densitypscore1$treat)
# after match
logitpscore1 <- c(match_ATT1$mdata$X[1,],match_ATT1$mdata$X[2,])
xx<- vector(length=688)
for(i in 1:344){
xx[i] <- 1
xx[i+344] <- 0
}
densitypscore2 <- data.frame(pscore=logitpscore1,treat=xx)
sm.density.compare(densitypscore2$pscore,group = densitypscore2$treat)
##propensity score matching using new_cps
#ATT
logit_m2 <- glm(treat~age+education+black+hispanic+married+nodegree+re74+re75,data=new_cps,family = binomial(link = "logit"))
match_ATT2 <- Match(Y=new_cps$re78,Tr=new_cps$treat,X=logit_m2$fitted,estimand="ATT",M=1,replace=T)
logit_m3 <- glm(treat~age+education+black+hispanic+married+nodegree+re74+re75+I(married*re75)+I(age*re74),data=new_cps,family = binomial(link = "logit"))
match_ATT3 <- Match(Y=new_cps$re78,Tr=new_cps$treat,X=logit_m3$fitted,estimand="ATT",M=1,replace=T)
summary(match_ATT2 )
summary(match_ATT3 )
# Cov Balance test
mb1 <- MatchBalance(treat~age+education+black+hispanic+married+nodegree+re74+re75,
data=new_cps, match.out=match_ATT2, nboots=10)
mb2 <- MatchBalance(treat~age+education+black+hispanic+married+nodegree+re74+re75+I(married*re75)+I(age*re74),
data=new_cps, match.out=match_ATT3, nboots=10)
# use package matchit to update data after match, DATA nsw_dw
library(MatchIt)
res_nsw_dw <- matchit(treat~age+education+black+black+hispanic+married+nodegree+re74+re75,
data=nsw_dw,method="nearest",ratio=1) # matching
library(Zelig)
z.out <- zelig(re78~treat+age+education+black+black+hispanic+married+nodegree+re74+re75,data=nsw_dw,model="ls")
xtreat <- setx(z.out,treat=1)
xcontrol <- setx(z.out,treat=0)
ATE <- sim(z.out,x=xcontrol,x1=xtreat)
summary(ATE)
res_nsw_dwA <- match.data(res_nsw_dw) # ALL DATA after match
res_nsw_dwT <- match.data(res_nsw_dw,group = "treat") # Treated group
res_nsw_dwC <- match.data(res_nsw_dw,group = "control") # Controls group after match
#-------------------------------------#
#-------Chapter7: IV estimation-------#
#-------------------------------------#
library(AER) # package created by Christian Kleiber & Achim Zeileis
# the authors of 《Applied Econometrics with R》
angrist <- read.dta13("ch7\\angrist.dta")
iv_fit_QTR1 <- ivreg(LWKLYWGE~EDUC|QTR1,data=angrist) #iv is QTR1 and result in P142
# first stage reg by ols:
# iv_fit_QTR1_freg <- lm(EDUC~QTR1,data=angrist)
# summary(iv_fit_QTR1_freg)
# first table in the p142
iv_fit_QTR4 <- ivreg(LWKLYWGE~EDUC|QTR4,data=angrist) #iv is QTR4 and result in P142
iv_fit_QTR123 <- ivreg(LWKLYWGE~EDUC|QTR1+QTR2+QTR3,data=angrist) #iv is QTR1,QTR2,QTR3
summary(iv_fit_QTR123) #result in P143
# Estimation of 7.3
angrist$year <- angrist$YOB+(angrist$QOB-1)/4
plotdata <- angrist[,c(65,3,6)]
plotdata_new <- ddply(plotdata,.(year),colwise(mean))
plot(plotdata_new$year,plotdata_new$EDUC,type = "b") # figure 7.2
plot(plotdata_new$year,plotdata_new$LWKLYWGE,type = "b") # figure 7.3
# another method for graphing
# library(ggplot2)
# ggplot(data=plotdata_new,aes(year,EDUC))+geom_point(colour=year)
## REG discrebed in Table in P145
# Colume 1、3、5、7
colume1_OLS <- lm(LWKLYWGE~EDUC+YR20+YR21+YR22+YR23+YR24+YR25+YR26+YR27+YR28,data=angrist)
colume3_OLS <- lm(LWKLYWGE~EDUC+AGEQ+AGEQSQ
+YR20+YR21+YR22+YR23+YR24+YR25+YR26+YR27+YR28,data=angrist)
colume5_OLS <- lm(LWKLYWGE~EDUC+RACE+MARRIED+SMSA
+NEWENG+MIDATL+ESOCENT+WNOCENT+SOATL+ESOCENT+WSOCENT+MT
+YR20+YR21+YR22+YR23+YR24+YR25+YR26+YR27+YR28,data=angrist)
colume7_OLS <- lm(LWKLYWGE~EDUC+AGEQ+AGEQSQ+RACE+MARRIED+SMSA
+NEWENG+MIDATL+ESOCENT+WNOCENT+SOATL+ESOCENT+WSOCENT+MT
+YR20+YR21+YR22+YR23+YR24+YR25+YR26+YR27+YR28,data=angrist)
# Colume 2、4、6、8
colume2_iv<- ivreg(LWKLYWGE~EDUC
+YR20+YR21+YR22+YR23+YR24+YR25+YR26+YR27+YR28|.-EDUC #multi-variablies,claim the edengous EDUC
+QTR120+QTR121+QTR122+QTR123+QTR124+QTR125+QTR126+QTR127
+QTR128+QTR129+QTR220+QTR221+QTR222+QTR223+QTR224+QTR225
+QTR226+QTR227+QTR228+QTR229+QTR320+QTR321+QTR322+QTR323
+QTR324+QTR325+QTR326+QTR327+QTR328+QTR329+YR20+YR21+YR22
+YR23+YR24+YR25+YR26+YR27+YR28,data=angrist)
colume4_iv<- ivreg(LWKLYWGE~EDUC+AGEQ+AGEQSQ
+YR20+YR21+YR22+YR23+YR24+YR25+YR26+YR27+YR28|.-EDUC
+QTR120+QTR121+QTR122+QTR123+QTR124+QTR125+QTR126+QTR127
+QTR128+QTR129+QTR220+QTR221+QTR222+QTR223+QTR224+QTR225
+QTR226+QTR227+QTR228+QTR229+QTR320+QTR321+QTR322+QTR323
+QTR324+QTR325+QTR326+QTR327+QTR328+QTR329+YR20+YR21+YR22
+YR23+YR24+YR25+YR26+YR27+YR28,data=angrist)
colume6_iv<- ivreg(LWKLYWGE~EDUC+RACE+MARRIED+SMSA
+NEWENG+MIDATL+ESOCENT+WNOCENT+SOATL+ESOCENT+WSOCENT+MT
+YR20+YR21+YR22+YR23+YR24+YR25+YR26+YR27+YR28|.-EDUC
+QTR120+QTR121+QTR122+QTR123+QTR124+QTR125+QTR126+QTR127
+QTR128+QTR129+QTR220+QTR221+QTR222+QTR223+QTR224+QTR225
+QTR226+QTR227+QTR228+QTR229+QTR320+QTR321+QTR322+QTR323
+QTR324+QTR325+QTR326+QTR327+QTR328+QTR329+YR20+YR21+YR22
+YR23+YR24+YR25+YR26+YR27+YR28,data=angrist)
colume8_iv<- ivreg(LWKLYWGE~EDUC+AGEQ+AGEQSQ+RACE+MARRIED+SMSA
+NEWENG+MIDATL+ESOCENT+WNOCENT+SOATL+ESOCENT+WSOCENT+MT
+YR20+YR21+YR22+YR23+YR24+YR25+YR26+YR27+YR28|.-EDUC
+QTR120+QTR121+QTR122+QTR123+QTR124+QTR125+QTR126+QTR127
+QTR128+QTR129+QTR220+QTR221+QTR222+QTR223+QTR224+QTR225
+QTR226+QTR227+QTR228+QTR229+QTR320+QTR321+QTR322+QTR323
+QTR324+QTR325+QTR326+QTR327+QTR328+QTR329+YR20+YR21+YR22
+YR23+YR24+YR25+YR26+YR27+YR28,data=angrist)
#---------------------------------------------#
#-------Chapter8: Mehtods of Panel Data-------#
#---------------------------------------------#
# DiD
cardkrueger <- read.dta13("ch8\\cardkrueger1994.dta")
cardkrueger$treat <- ifelse(cardkrueger$treated=="NJ",cardkrueger$treat <- 1,cardkrueger$treat <- 0)
Did <- lm(fte~treat+t+I(treat*t),data=cardkrueger)
summary(Did) # or Did$coefficients
# Synth
library(Synth) # Package created by Jens Hainmueller and Alexis Diamond
smoking <- read.dta13("ch8\\smoking.dta")
smoking$id <- as.numeric(smoking$state)
smoking$name <- as.character(smoking$state)
smoke_synth <- dataprep(
foo = smoking,
predictors = c("lnincome","age15to24","retprice"), # set other variables which time ranging is 1980:1988
predictors.op = "mean",
dependent = "cigsale",
unit.variable = "id",
time.variable = "year",
special.predictors = list(
list("cigsale", c(1975,1980,1988), "mean"), # set vatiables which time ranging is different from 1980:1988
list("beer", c(1984:1988), "mean")),
treatment.identifier = 3, # id of treatment
controls.identifier = c(1,2,4:39),
time.predictors.prior =c(1980:1988),
time.optimize.ssr = c(1970:1987), # MSPE
unit.names.variable = "name",
time.plot = 1970:2000 # Time
)
synth.out <- synth(data.prep.obj = smoke_synth)
synth.tables <- synth.tab(
dataprep.res = smoke_synth,
synth.res = synth.out
)
# results tables:
print(synth.tables)
# plot results:
# path
path.plot(synth.res = synth.out,
dataprep.res = smoke_synth,
Legend = c("Basque country","synthetic Basque country"))
## gaps
gaps.plot(synth.res = synth.out,
dataprep.res = smoke_synth)
#####NoteS: the result is differnt from the book . I do not know the reason.
#-------------------------------------------------#
#-------Chapter9: Regression Discontinuity--------#
#-------------------------------------------------#
library(rdrobust)
lee <- read.dta13("ch9\\lee.dta")
# Replicat the Result in Page220
Lee_1 <- lee[lee$margin >= -0.28 & lee$margin <= 0.28,]
Lee_1$dd <- ifelse(Lee_1$margin>0,Lee_1$d <- 1,Lee_1$d <- 0) # Creat Rdd
summary(lm(vote~dd,data=Lee_1)) #Result of Colume1
summary(lm(vote~dd+margin+I(dd*margin),data=Lee_1)) #Result in Colume2
# Replicat the Result in Page221
Lee_2 <- lee[lee$margin >= -0.5 & lee$margin <= 0.5,]
Lee_2$dd <- ifelse(Lee_2$margin>0,Lee_2$d <- 1,Lee_2$d <- 0) # Creat Rdd
summary(lm(vote~dd+margin+I(dd*margin),data=Lee_2)) # Result of Colume1
summary(lm(vote~dd+margin+I(margin^2)+I(dd*margin)+I(dd*margin^2),data=Lee_2)) # Result of Colume2
summary(lm(vote~dd+margin+I(margin^2)+I(margin^3)+I(dd*margin)+I(dd*margin^2)+I(dd*margin^3),data=Lee_2)) # Result of Colume3
summary(lm(vote~dd+margin+I(margin^2)+I(margin^3)+I(dd*margin)+I(dd*margin^2)+I(dd*margin^3)+I(dd*margin^4)+I(dd*margin^4),data=Lee_2)) # Result of Colume4
# Figure 9.12 using data Lee_2 in Page218
rdplot(y = Lee_2$vote, x = Lee_2$margin, nbins=50, p=2, # Defult is p=4
x.lim=c(-0.5,0.5),
y.lim=c(0.2,0.8),col.lines ="black",
x.label = 'Democratic Vote Share Margin of Victory, Election t',
y.label = 'Vote Share, Election t+1')
# Figure 9.13 in Page223
Lee_2_test1 <- Lee_2[Lee_2$margin < 0,]
Lee_2_test2 <- Lee_2[Lee_2$margin >= 0,]
rdplot(y = Lee_2_test1$vote, x = Lee_2_test1$margin, c=-0.25,p=4,
x.lim=c(-0.5,0),y.lim=c(0.2,0.5),
x.label=c("断点为-0.25"),col.lines ="black")
rdplot(y = Lee_2_test2$vote, x = Lee_2_test2$margin, c=0.25,p=4,
x.lim=c(0,0.5),y.lim=c(0.5,0.75),
x.label=c("断点为0.25"),col.lines ="black")
# Figure 9.8 in Page 206 and code edited by stata in Page219.
rdbwselect_2014(y = Lee_2$vote, x = Lee_2$margin, c=0,
kernel = "uniform", bwselect = "CV",
cvgrid_min = 0.1, cvgrid_max = 0.5, cvplot = TRUE)
# use rdbwselect_2014 rather rdbwselect
# Result shows the best bindwith is 0.28
# Result in Page226.
summary(rdbwselect(y = Lee_2$vote, x =Lee_2$margin, kernel = "uniform", c=0, all = T))
# Replicat the Result in Page227
summary(rdrobust(y = Lee_2$vote, x = Lee_2$margin,c=0, kernel = "uniform", all=T))
# Replicat the Result in Page228
set.seed(123)
x <- rnorm(10000,mean=10,sd=2)
e <- rnorm(10000,mean=0,sd=1)
u <- rnorm(10000,mean=0,sd=1)
for(i in 1:10000){
D[i] <- min(0.5*x[i]+u[i],5+u[i])
Y[i] <- 2*D[i]-0.5*x[i]+e[i]
}
rdplot(y = D, x = x, c=10, p=2,x.lim=c(0,20),y.lim=c(0,6),col.lines ="black",
x.label = "x",y.label = "D") # Figure 228 a
rdplot(y = Y, x = x, c=10, p=2,x.lim=c(0,20),y.lim=c(-2,6),col.lines ="black",
x.label = "x",y.label = "Y") # Figure 228 b
summary(rdrobust(Y, x, c=10, kernel = "uniform", fuzzy=D, deriv=1, p=2, all=T))
#-------------------------------------------------#
#-------Chapter10: Bunching-----------------------#
#-------------------------------------------------#
library(bunchr)
##-----------------------------------------------------Analyzing a kink
set.seed(42)
ability_vec <- 4000 * rbeta(100000, 2, 5) #生成模拟数
earning_vec <- sapply(ability_vec, earning_fun, 0.2, 0, 0.1, 0, 1000)
# earning_fun(n, elas, t1, t2, Tax, zstar)
# Arguments
#
# n Ability of person (earnings with zero tax)
# elas elasticity of earnings w.r.t. net-of-tax rate
# t1 Tax rate before notch/kink
# t2 Tax rate after notch/kink
# Tax height of notch (zero for pure kink)
# zstar place of notch/kink (critical earning point)
bunch_viewer(earning_vec, 1000, cf_start = 10, cf_end = 10, exclude_before = 2,
exclude_after = 2, binw = 50, trimy=F) #图示
kink_est1 <- bunch(earning_vec, zstar = 1000, t1 = 0, t2 = 0.1, Tax = 0,
cf_start = 10, cf_end = 10,
exclude_before = 2, exclude_after = 2, binw = 50, nboots = 100) ##已自动生成反事实数据
# Usage
#
# bunch(earnings, zstar, t1, t2, Tax = 0, cf_start = NA, cf_end = NA,
# exclude_before = NA, exclude_after = NA, force_after = FALSE,
# binw = 10, poly_size = 7, convergence = 0.01, max_iter = 100,
# correct = TRUE, select = TRUE, draw = TRUE, nboots = 0, seed = NA,
# progress = FALSE, title = "Bunching Visualization",
# varname = "Earnings")
# Arguments
#
# earnings Vector of earnings, hopefully a very large one.
# zstar Place of kink (critical earning point).
# t1 Marginal tax rate before kink.
# t2 Marginal tax rate after kink.
# Tax "Penalty" tax for crossing zstar.
# cf_start Number of bins before the kink bin where counter-factual histogram should start.
# cf_end Number of bins after the kink bin where counter-factual histogram should start.
# exclude_before Number of excluded bins before the kink bin.
# exclude_after Number of excluded bins after the kink bin.
# force_after For notch analysis, should bunch be forced to use of the provided exclude_after for the end of the bunching, rather than trying to find the bin where the sum of the integral is zero? See details at notch_estimator documentation.
# binw Bin width.
# poly_size Order of polynomial used to calculate counter-factual histogram.
# convergence Minimal rate of change of bunching estimate to stop iterations.
# max_iter Maximum number of iterations for bunching estimates.
# correct Should the counter-factual histogram be corrected to compensate for shifting left because of the notch? See details.
# select Should model selection be used to find counter-factual histogram? See details.
# draw Should a graph be drawn?
# nboots how many bootstraps should be run?
# seed specify seed for bootstraps (earnings sampling).
# progress Should a progress bar be desplayed?
# title Title for Plot output
# varname Name for running variable, to be desplayed in the plot
kink_est1$e
## [1] 0.186233
quantile(kink_est1$booted_e, probs=c(0, 0.05, 0.1, 0.5, 0.9, 0.95, 1))
## 0% 5% 10% 50% 90% 95%
## 0.09330628 0.11666582 0.13313885 0.18567137 0.23956820 0.25130516
## 100%
## 0.32462423
kink_est1$Bn
## [1] 1170.216
quantile(kink_est1$booted_Bn, probs=c(0, 0.05, 0.1, 0.5, 0.9, 0.95, 1))
## 0% 5% 10% 50% 90% 95% 100%
## 599.0476 744.2115 851.2940 1166.8193 1481.8489 1541.6775 1939.3720
#show Counter-Factuals
kink_est1$data$cf_counts
###调整带宽
kink_est2 <- bunch(earning_vec, zstar = 1000, t1 = 0, t2 = 0.1, Tax = 0,
cf_start = 500, cf_end = 500,
exclude_before = 10, exclude_after = 10, binw = 1,
nboots = 100, seed = 123, draw = F)
kink_est2$e
## [1] 0.1932915
quantile(kink_est2$booted_e, probs=c(0, 0.05, 0.1, 0.5, 0.9, 0.95, 1))
## 0% 5% 10% 50% 90% 95% 100%
## 0.1770582 0.1802953 0.1834727 0.1915006 0.2035105 0.2061229 0.2116499
kink_est2$Bn
## [1] 1223.928
quantile(kink_est2$booted_Bn, probs=c(0, 0.05, 0.1, 0.5, 0.9, 0.95, 1))
## 0% 5% 10% 50% 90% 95% 100%
## 1127.625 1143.042 1164.198 1219.175 1284.605 1304.872 1332.322
#show Counter-Factuals
kink_est2$data$cf_counts
##-----------------------------------------------------Analyzing a notch
earning_vec <- sapply(ability_vec, earning_fun, 0.2, 0.1, 0.1, 200, 1000)
bunch_viewer(earning_vec, 1000, 20, 50, 2, 25, binw = 20)
notch_est <- bunch(earning_vec, zstar = 1000, t1 = 0.1, t2 = 0.1, Tax = 200,
cf_start = 20, cf_end = 50, force_after = FALSE,
exclude_before = 2, exclude_after = 25, binw = 20,
nboots = 100, seed = 123)
notch_est$e
## [1] 0.1843316
quantile(notch_est$booted_e, probs=c(0, 0.05, 0.1, 0.5, 0.9, 0.95, 1))
## 0% 5% 10% 50% 90% 95% 100%
## 0.1203677 0.1407937 0.1621033 0.2074509 0.2074509 0.2074509 0.2074509
ability_vec_small <- 4000 * rbeta(10000, 2, 5)
earnings_small <- sapply(ability_vec_small, earning_fun, 0.2, 0, 0.1, 0, 1000)
kink_est <- bunch(earnings_small, zstar = 1000, t1 = 0, t2 = 0.1, Tax = 0,
cf_start = 10, cf_end = 10,
exclude_before = 1, exclude_after = 1, binw = 50,
nboots = 100, seed = 123)
kink_est$e
## [1] 0.1662341
quantile(kink_est$booted_e, probs=c(0, 0.05, 0.1, 0.5, 0.9, 0.95, 1))
## 0% 5% 10% 50% 90% 95%
## 0.04387312 0.07718377 0.09195058 0.17941217 0.28779019 0.30870107
## 100%
## 0.39275768
##大样本
ability_vec_large <- 4000 * rbeta(1000000, 2, 5)
earnings_large <- sapply(ability_vec_large, earning_fun, 0.2, 0, 0.1, 0, 1000)
kink_est <- bunch(earnings_large, zstar = 1000, t1 = 0, t2 = 0.1, Tax = 0,
cf_start = 50, cf_end = 50,
exclude_before = 5, exclude_after = 5, binw = 10,
nboots = 100, seed = 123)
kink_est$e
## [1] 0.183655
quantile(kink_est$booted_e, probs=c(0, 0.05, 0.1, 0.5, 0.9, 0.95, 1))
## 0% 5% 10% 50% 90% 95% 100%
## 0.1688440 0.1742967 0.1772704 0.1850824 0.1928552 0.1942905 0.1970645
kink_est <- bunch(earnings_large, zstar = 1000, t1 = 0, t2 = 0.1, Tax = 0,
cf_start = 50, cf_end = 50,
exclude_before = 1, exclude_after = 1, binw = 5,
nboots = 100, seed = 123, draw = F)
kink_est$e
## [1] 0.1949573
quantile(kink_est$booted_e, probs=c(0, 0.05, 0.1, 0.5, 0.9, 0.95, 1))
## 0% 5% 10% 50% 90% 95% 100%
## 0.1879073 0.1901380 0.1915570 0.1955756 0.1988548 0.2003031 0.2025746
#---------------------------------#
#-------Important REFERENCE-------#
#---------------------------------#
# [1]Abadie A, Diamond A, Hainmueller J. Synth: An R Package for Synthetic Control Methods in Comparative Case Studies[J]. Journal of Statistical Software, 2011, 42(13):1-17.
# [2]Calonico, S., M. D. Cattaneo, and R. Titiunik. rdrobust: An R Package for Robust Nonparametric Inference in Regression-Discontinuity Designs. R Journal, 2015b, 7(1): 38-51.