load("~/Downloads/datamatch.RData")
head(datamatch)
## polling.place EV age.group educ male tech
## 1 Escuela de Comercio Juan Manuel de Estrada 1 4 4 1 2
## 2 Escuela de Comercio Juan Manuel de Estrada 1 2 3 1 2
## 3 Escuela de Comercio Juan Manuel de Estrada 1 4 1 1 2
## 4 Escuela de Comercio Juan Manuel de Estrada 1 3 7 1 3
## 5 Escuela de Comercio Juan Manuel de Estrada 1 4 3 1 3
## 6 Escuela de Comercio Juan Manuel de Estrada 1 3 2 0 3
## pol.info white.collar not.full.time capable.auth eval.voting easy.voting
## 1 2 0 0 1 1 0
## 2 1 0 1 0 1 0
## 3 1 1 0 1 0 0
## 4 3 0 0 0 1 0
## 5 2 0 0 1 0 0
## 6 1 0 1 1 0 0
## sure.counted conf.secret how.clean speed agree.evoting eselect.cand
## 1 1 1 1 1 1 1
## 2 1 0 NA 1 1 1
## 3 1 1 0 1 0 1
## 4 0 1 0 1 1 1
## 5 1 1 1 1 1 1
## 6 NA 0 0 1 1 1
# 1502 observations
# load necessary libraries
library(MatchIt)
library(Matching)
## Loading required package: MASS
## ##
## ## Matching (Version 4.9-11, Build Date: 2021-10-18)
## ## See http://sekhon.berkeley.edu/matching for additional documentation.
## ## Please cite software as:
## ## Jasjeet S. Sekhon. 2011. ``Multivariate and Propensity Score Matching
## ## Software with Automated Balance Optimization: The Matching package for R.''
## ## Journal of Statistical Software, 42(7): 1-52.
## ##
library(rgenoud)
## ## rgenoud (Version 5.8-3.0, Build Date: 2019-01-22)
## ## See http://sekhon.berkeley.edu/rgenoud for additional documentation.
## ## Please cite software as:
## ## Walter Mebane, Jr. and Jasjeet S. Sekhon. 2011.
## ## ``Genetic Optimization Using Derivatives: The rgenoud package for R.''
## ## Journal of Statistical Software, 42(11): 1-26.
## ##
# only get necessary columns - covariates + treatment + 1 outcome
data_new <- datamatch[, c(2, 3, 4, 5, 6, 7, 8, 9, 17)]
head(data_new)
## EV age.group educ male tech pol.info white.collar not.full.time agree.evoting
## 1 1 4 4 1 2 2 0 0 1
## 2 1 2 3 1 2 1 0 1 1
## 3 1 4 1 1 2 1 1 0 0
## 4 1 3 7 1 3 3 0 0 1
## 5 1 4 3 1 3 2 0 0 1
## 6 1 3 2 0 3 1 0 1 1
# drop missing values
data_omit <- na.omit(data_new)
head(data_omit)
## EV age.group educ male tech pol.info white.collar not.full.time agree.evoting
## 1 1 4 4 1 2 2 0 0 1
## 2 1 2 3 1 2 1 0 1 1
## 3 1 4 1 1 2 1 1 0 0
## 4 1 3 7 1 3 3 0 0 1
## 5 1 4 3 1 3 2 0 0 1
## 6 1 3 2 0 3 1 0 1 1
# 1409 observations
### REPLICATION: Table 2 - Pre-Matching
datamatch[, 10:18][is.na(datamatch[, 10:18]) == "TRUE"] <- 99999
datamatch <- na.omit(datamatch)
EV <- datamatch[2]
covariates <- datamatch[c("age.group", "educ", "white.collar", "not.full.time", "male", "tech", "pol.info")]
covariate.lbls <- names(covariates)
n.covariates <- dim(covariates)[2]
tab2.pre <- matrix(NA, nrow = n.covariates, ncol = 4)
rownames(tab2.pre) <- covariate.lbls
colnames(tab2.pre) <- c("ev", "tv", "diff", "pvalue")
tab2.pre[, 1:2] <- cbind(apply(covariates[EV == 1,], 2, mean), apply(covariates[EV == 0,], 2, mean))
tab2.pre[, 3] <- tab2.pre[, 1] - tab2.pre[, 2]
for (i in c(1, 2, 6, 7)){
tab2.pre[i, 4] <- ks.boot(covariates[, i][EV == 1], covariates[, i][EV == 0], nboots = 500)$ks.boot.pvalue
}
for (i in c(3, 4, 5)){
tab2.pre[i, 4] <- prop.test(table(covariates[, i], EV$EV), n = apply(table(covariates[,i],EV$EV),2, sum))$p.value
}
print(tab2.pre)
## ev tv diff pvalue
## age.group 2.4757506 2.4433498 0.032400824 0.54600000
## educ 4.7713626 4.1428571 0.628505444 0.00000000
## white.collar 0.3025404 0.2758621 0.026678347 0.29287524
## not.full.time 0.2771363 0.3349754 -0.057839111 0.01998267
## male 0.4965358 0.4909688 0.005566995 0.87472467
## tech 4.1836028 3.9096880 0.273914758 0.00000000
## pol.info 1.4745958 1.3103448 0.164251015 0.00000000
# REPLICATION: Propensity Score Matching
# logistical regression - propensity scores
logit <- glm(EV ~ age.group + I(age.group^2) + I(age.group^3) + age.group:educ + age.group:tech + educ + I(educ^2) + tech + I(tech^2) + pol.info + educ:pol.info + age.group:pol.info + tech:pol.info + white.collar + not.full.time + male, data = datamatch, family = "binomial")
# matching
X <- logit$fitted
Y <- datamatch$agree.evoting
Tr <- datamatch$EV
rr0 <- Match(Y=Y, Tr=Tr, X=X, caliper = 0.05)
# check covariate balance
mb_psm0 <- MatchBalance(EV ~ age.group + educ + white.collar + not.full.time + male + tech + pol.info, data = datamatch, match.out = rr0)
##
## ***** (V1) age.group *****
## Before Matching After Matching
## mean treatment........ 2.4758 2.4713
## mean control.......... 2.4433 2.4289
## std mean diff......... 2.4035 3.1461
##
## mean raw eQQ diff..... 0.065681 0.040388
## med raw eQQ diff..... 0 0
## max raw eQQ diff..... 1 1
##
## mean eCDF diff........ 0.013296 0.0080777
## med eCDF diff........ 0.017038 0.0062136
## max eCDF diff........ 0.026754 0.017864
##
## var ratio (Tr/Co)..... 1.048 1.0415
## T-test p-value........ 0.64508 0.38493
## KS Bootstrap p-value.. 0.538 0.346
## KS Naive p-value...... 0.96005 0.80582
## KS Statistic.......... 0.026754 0.017864
##
##
## ***** (V2) educ *****
## Before Matching After Matching
## mean treatment........ 4.7714 4.7544
## mean control.......... 4.1429 4.7574
## std mean diff......... 27.294 -0.13037
##
## mean raw eQQ diff..... 0.62397 0.050874
## med raw eQQ diff..... 0 0
## max raw eQQ diff..... 2 2
##
## mean eCDF diff........ 0.070303 0.0063107
## med eCDF diff........ 0.085505 0.0013592
## max eCDF diff........ 0.1307 0.018641
##
## var ratio (Tr/Co)..... 1.3199 1.0132
## T-test p-value........ 3.0007e-08 0.9616
## KS Bootstrap p-value.. < 2.22e-16 0.372
## KS Naive p-value...... 9.9039e-06 0.76224
## KS Statistic.......... 0.1307 0.018641
##
##
## ***** (V3) white.collar *****
## Before Matching After Matching
## mean treatment........ 0.30254 0.30526
## mean control.......... 0.27586 0.29773
## std mean diff......... 5.8044 1.6338
##
## mean raw eQQ diff..... 0.026273 0.0066019
## med raw eQQ diff..... 0 0
## max raw eQQ diff..... 1 1
##
## mean eCDF diff........ 0.013339 0.003301
## med eCDF diff........ 0.013339 0.003301
## max eCDF diff........ 0.026678 0.0066019
##
## var ratio (Tr/Co)..... 1.0558 1.0143
## T-test p-value........ 0.26506 0.67536
##
##
## ***** (V4) not.full.time *****
## Before Matching After Matching
## mean treatment........ 0.27714 0.27485
## mean control.......... 0.33498 0.25578
## std mean diff......... -12.915 4.269
##
## mean raw eQQ diff..... 0.059113 0.020194
## med raw eQQ diff..... 0 0
## max raw eQQ diff..... 1 1
##
## mean eCDF diff........ 0.02892 0.010097
## med eCDF diff........ 0.02892 0.010097
## max eCDF diff........ 0.057839 0.020194
##
## var ratio (Tr/Co)..... 0.89885 1.047
## T-test p-value........ 0.018169 0.2582
##
##
## ***** (V5) male *****
## Before Matching After Matching
## mean treatment........ 0.49654 0.4924
## mean control.......... 0.49097 0.47806
## std mean diff......... 1.1128 2.8658
##
## mean raw eQQ diff..... 0.0049261 0.0007767
## med raw eQQ diff..... 0 0
## max raw eQQ diff..... 1 1
##
## mean eCDF diff........ 0.0027835 0.00038835
## med eCDF diff........ 0.0027835 0.00038835
## max eCDF diff........ 0.005567 0.0007767
##
## var ratio (Tr/Co)..... 0.99979 1.0017
## T-test p-value........ 0.83338 0.4814
##
##
## ***** (V6) tech *****
## Before Matching After Matching
## mean treatment........ 4.1836 4.1906
## mean control.......... 3.9097 4.2089
## std mean diff......... 16.549 -1.1048
##
## mean raw eQQ diff..... 0.26929 0.04932
## med raw eQQ diff..... 0 0
## max raw eQQ diff..... 1 1
##
## mean eCDF diff........ 0.045652 0.0082201
## med eCDF diff........ 0.041212 0.0054369
## max eCDF diff........ 0.10917 0.024854
##
## var ratio (Tr/Co)..... 1.0529 1.0165
## T-test p-value........ 0.0015253 0.75787
## KS Bootstrap p-value.. < 2.22e-16 0.124
## KS Naive p-value...... 0.0003975 0.40413
## KS Statistic.......... 0.10917 0.024854
##
##
## ***** (V7) pol.info *****
## Before Matching After Matching
## mean treatment........ 1.4746 1.4713
## mean control.......... 1.3103 1.4336
## std mean diff......... 20.451 4.719
##
## mean raw eQQ diff..... 0.16256 0.025243
## med raw eQQ diff..... 0 0
## max raw eQQ diff..... 1 1
##
## mean eCDF diff........ 0.041063 0.0063107
## med eCDF diff........ 0.035125 0.0021359
## max eCDF diff........ 0.094002 0.020971
##
## var ratio (Tr/Co)..... 1.44 1.0445
## T-test p-value........ 2.0928e-05 0.13163
## KS Bootstrap p-value.. < 2.22e-16 0.032
## KS Naive p-value...... 0.0036036 0.62301
## KS Statistic.......... 0.094002 0.020971
##
##
## Before Matching Minimum p.value: < 2.22e-16
## Variable Name(s): educ tech pol.info Number(s): 2 6 7
##
## After Matching Minimum p.value: 0.032
## Variable Name(s): pol.info Number(s): 7
summary(rr0)
##
## Estimate... 483.32
## AI SE...... 1348.4
## T-stat..... 0.35845
## p.val...... 0.72001
##
## Original number of observations.............. 1475
## Original number of treated obs............... 866
## Matched number of observations............... 855
## Matched number of observations (unweighted). 2575
##
## Caliper (SDs)........................................ 0.05
## Number of obs dropped by 'exact' or 'caliper' 11
# REPEAT with simpler model! And newly cleaned data!
# logistical regression - propensity scores
logit1 <- glm(EV ~ age.group + educ + tech + pol.info + white.collar + not.full.time + male, data = data_omit, family = "binomial")
# matching
X1 <- logit1$fitted
Y1 <- data_omit$agree.evoting
Tr1 <- data_omit$EV
rr1 <- Match(Y=Y1, Tr=Tr1, X=X1, caliper = 0.01, BiasAdjust = TRUE)
# check covariate balance
mb_psm1 <- MatchBalance(EV ~ age.group + educ + white.collar + not.full.time + male + tech + pol.info, data = data_omit, match.out = rr1, nboots = 2000)
##
## ***** (V1) age.group *****
## Before Matching After Matching
## mean treatment........ 2.4673 2.435
## mean control.......... 2.446 2.4645
## std mean diff......... 1.5948 -2.1903
##
## mean raw eQQ diff..... 0.063465 0.038749
## med raw eQQ diff..... 0 0
## max raw eQQ diff..... 1 1
##
## mean eCDF diff........ 0.012531 0.0077498
## med eCDF diff........ 0.017628 0
## max eCDF diff........ 0.024371 0.019608
##
## var ratio (Tr/Co)..... 1.0361 0.97536
## T-test p-value........ 0.76583 0.61124
## KS Bootstrap p-value.. 0.6615 0.3485
## KS Naive p-value...... 0.98724 0.80476
## KS Statistic.......... 0.024371 0.019608
##
##
## ***** (V2) educ *****
## Before Matching After Matching
## mean treatment........ 4.7615 4.3771
## mean control.......... 4.0909 4.3839
## std mean diff......... 29.258 -0.32584
##
## mean raw eQQ diff..... 0.66724 0.0831
## med raw eQQ diff..... 1 0
## max raw eQQ diff..... 2 2
##
## mean eCDF diff........ 0.075703 0.010154
## med eCDF diff........ 0.08996 0.0065359
## max eCDF diff........ 0.14546 0.026611
##
## var ratio (Tr/Co)..... 1.3492 1.0394
## T-test p-value........ 5.3884e-09 0.88181
## KS Bootstrap p-value.. < 2.22e-16 0.1535
## KS Naive p-value...... 1.047e-06 0.43419
## KS Statistic.......... 0.14546 0.026611
##
##
## ***** (V3) white.collar *****
## Before Matching After Matching
## mean treatment........ 0.30508 0.29661
## mean control.......... 0.2813 0.30149
## std mean diff......... 5.1617 -1.0673
##
## mean raw eQQ diff..... 0.024014 0.020542
## med raw eQQ diff..... 0 0
## max raw eQQ diff..... 1 1
##
## mean eCDF diff........ 0.011891 0.010271
## med eCDF diff........ 0.011891 0.010271
## max eCDF diff........ 0.023781 0.020542
##
## var ratio (Tr/Co)..... 1.0481 0.99069
## T-test p-value........ 0.33355 0.82806
##
##
## ***** (V4) not.full.time *****
## Before Matching After Matching
## mean treatment........ 0.27603 0.2839
## mean control.......... 0.33448 0.28468
## std mean diff......... -13.067 -0.17363
##
## mean raw eQQ diff..... 0.058319 0.0098039
## med raw eQQ diff..... 0 0
## max raw eQQ diff..... 1 1
##
## mean eCDF diff........ 0.029224 0.004902
## med eCDF diff........ 0.029224 0.004902
## max eCDF diff........ 0.058448 0.0098039
##
## var ratio (Tr/Co)..... 0.89728 0.99834
## T-test p-value........ 0.019525 0.96409
##
##
## ***** (V5) male *****
## Before Matching After Matching
## mean treatment........ 0.49879 0.48729
## mean control.......... 0.49571 0.50022
## std mean diff......... 0.61513 -2.5857
##
## mean raw eQQ diff..... 0.0017153 0.011204
## med raw eQQ diff..... 0 0
## max raw eQQ diff..... 1 1
##
## mean eCDF diff........ 0.0015388 0.0056022
## med eCDF diff........ 0.0015388 0.0056022
## max eCDF diff........ 0.0030775 0.011204
##
## var ratio (Tr/Co)..... 0.99956 0.99935
## T-test p-value........ 0.90949 0.56068
##
##
## ***** (V6) tech *****
## Before Matching After Matching
## mean treatment........ 4.1949 4.0636
## mean control.......... 3.916 4.0685
## std mean diff......... 17.053 -0.3004
##
## mean raw eQQ diff..... 0.27616 0.045285
## med raw eQQ diff..... 0 0
## max raw eQQ diff..... 1 1
##
## mean eCDF diff........ 0.046494 0.0075475
## med eCDF diff........ 0.044928 0.0051354
## max eCDF diff........ 0.1073 0.024743
##
## var ratio (Tr/Co)..... 1.0308 1.0023
## T-test p-value........ 0.0015052 0.93936
## KS Bootstrap p-value.. < 2.22e-16 0.2005
## KS Naive p-value...... 0.00076398 0.52836
## KS Statistic.......... 0.1073 0.024743
##
##
## ***** (V7) pol.info *****
## Before Matching After Matching
## mean treatment........ 1.4685 1.3376
## mean control.......... 1.3053 1.3286
## std mean diff......... 20.407 1.3482
##
## mean raw eQQ diff..... 0.16295 0.020075
## med raw eQQ diff..... 0 0
## max raw eQQ diff..... 1 1
##
## mean eCDF diff........ 0.040801 0.0050187
## med eCDF diff........ 0.035003 0.0014006
## max eCDF diff........ 0.0932 0.017274
##
## var ratio (Tr/Co)..... 1.4448 1.0089
## T-test p-value........ 3.2741e-05 0.70849
## KS Bootstrap p-value.. < 2.22e-16 0.1385
## KS Naive p-value...... 0.0052778 0.90664
## KS Statistic.......... 0.0932 0.017274
##
##
## Before Matching Minimum p.value: < 2.22e-16
## Variable Name(s): educ tech pol.info Number(s): 2 6 7
##
## After Matching Minimum p.value: 0.1385
## Variable Name(s): pol.info Number(s): 7
summary(rr1)
##
## Estimate... 0.18273
## AI SE...... 0.023639
## T-stat..... 7.7301
## p.val...... 1.0658e-14
##
## Original number of observations.............. 1409
## Original number of treated obs............... 826
## Matched number of observations............... 708
## Matched number of observations (unweighted). 2142
##
## Caliper (SDs)........................................ 0.01
## Number of obs dropped by 'exact' or 'caliper' 118
# EXTENSION: Genetic Matching
set.seed(34664)
# the model like the original paper
genout <- GenMatch(Tr = data_omit$EV, X = cbind(data_omit$age.group, data_omit$educ, data_omit$male, data_omit$tech, data_omit$pol.info, data_omit$white.collar, data_omit$not.full.time), pop.size = 20, nboots = 500) # increase the number of bootstraps to increase the quality of matching
##
##
## Fri Apr 22 18:54:35 2022
## Domains:
## 0.000000e+00 <= X1 <= 1.000000e+03
## 0.000000e+00 <= X2 <= 1.000000e+03
## 0.000000e+00 <= X3 <= 1.000000e+03
## 0.000000e+00 <= X4 <= 1.000000e+03
## 0.000000e+00 <= X5 <= 1.000000e+03
## 0.000000e+00 <= X6 <= 1.000000e+03
## 0.000000e+00 <= X7 <= 1.000000e+03
##
## Data Type: Floating Point
## Operators (code number, name, population)
## (1) Cloning........................... 5
## (2) Uniform Mutation.................. 2
## (3) Boundary Mutation................. 2
## (4) Non-Uniform Mutation.............. 2
## (5) Polytope Crossover................ 2
## (6) Simple Crossover.................. 2
## (7) Whole Non-Uniform Mutation........ 2
## (8) Heuristic Crossover............... 2
## (9) Local-Minimum Crossover........... 0
##
## SOFT Maximum Number of Generations: 100
## Maximum Nonchanging Generations: 4
## Population size : 20
## Convergence Tolerance: 1.000000e-03
##
## Not Using the BFGS Derivative Based Optimizer on the Best Individual Each Generation.
## Not Checking Gradients before Stopping.
## Using Out of Bounds Individuals.
##
## Maximization Problem.
## GENERATION: 0 (initializing the population)
## Lexical Fit..... 1.045862e-01 1.045862e-01 2.400000e-01 3.173108e-01 3.173108e-01 3.173108e-01 3.230868e-01 8.720000e-01 8.953393e-01 9.597300e-01 9.660000e-01 1.000000e+00 1.000000e+00 1.000000e+00
## #unique......... 20, #Total UniqueCount: 20
## var 1:
## best............ 2.825860e+01
## mean............ 3.339641e+02
## variance........ 6.450582e+04
## var 2:
## best............ 8.035248e+02
## mean............ 4.941728e+02
## variance........ 8.645013e+04
## var 3:
## best............ 3.069706e+02
## mean............ 4.384428e+02
## variance........ 1.011329e+05
## var 4:
## best............ 8.780492e+02
## mean............ 4.598639e+02
## variance........ 9.014094e+04
## var 5:
## best............ 8.866684e+02
## mean............ 5.452365e+02
## variance........ 8.401963e+04
## var 6:
## best............ 4.663642e+02
## mean............ 3.831739e+02
## variance........ 9.354026e+04
## var 7:
## best............ 6.018247e+01
## mean............ 3.950401e+02
## variance........ 7.549191e+04
##
## GENERATION: 1
## Lexical Fit..... 2.960000e-01 3.173108e-01 3.173108e-01 3.701437e-01 3.701437e-01 4.120290e-01 5.271645e-01 5.271645e-01 5.827818e-01 8.737061e-01 9.660000e-01 9.980000e-01 1.000000e+00 1.000000e+00
## #unique......... 13, #Total UniqueCount: 33
## var 1:
## best............ 1.579388e+01
## mean............ 2.380219e+02
## variance........ 4.647217e+04
## var 2:
## best............ 8.068548e+02
## mean............ 5.581849e+02
## variance........ 6.920607e+04
## var 3:
## best............ 3.115044e+02
## mean............ 4.005003e+02
## variance........ 4.743333e+04
## var 4:
## best............ 8.926560e+02
## mean............ 4.773902e+02
## variance........ 1.139004e+05
## var 5:
## best............ 8.872365e+02
## mean............ 8.341435e+02
## variance........ 1.564155e+04
## var 6:
## best............ 4.693840e+02
## mean............ 3.829471e+02
## variance........ 5.323550e+04
## var 7:
## best............ 3.313543e+01
## mean............ 2.967744e+02
## variance........ 9.878792e+04
##
## GENERATION: 2
## Lexical Fit..... 3.060000e-01 3.173108e-01 3.173108e-01 3.701437e-01 3.701437e-01 4.120290e-01 5.271645e-01 5.271645e-01 5.827818e-01 8.737061e-01 9.660000e-01 9.960000e-01 1.000000e+00 1.000000e+00
## #unique......... 12, #Total UniqueCount: 45
## var 1:
## best............ 8.501244e+00
## mean............ 6.194932e+01
## variance........ 1.138634e+04
## var 2:
## best............ 8.068548e+02
## mean............ 7.565826e+02
## variance........ 1.653447e+04
## var 3:
## best............ 3.115044e+02
## mean............ 3.129946e+02
## variance........ 1.074162e+04
## var 4:
## best............ 8.926560e+02
## mean............ 6.971718e+02
## variance........ 7.259912e+04
## var 5:
## best............ 8.872365e+02
## mean............ 8.460005e+02
## variance........ 1.015081e+04
## var 6:
## best............ 4.693840e+02
## mean............ 4.181505e+02
## variance........ 2.155755e+04
## var 7:
## best............ 3.313543e+01
## mean............ 2.721703e+02
## variance........ 1.216237e+05
##
## GENERATION: 3
## Lexical Fit..... 3.173108e-01 3.173108e-01 3.300000e-01 3.701437e-01 3.701437e-01 4.120290e-01 5.271645e-01 5.271645e-01 5.827818e-01 8.737061e-01 9.720000e-01 9.980000e-01 1.000000e+00 1.000000e+00
## #unique......... 10, #Total UniqueCount: 55
## var 1:
## best............ 1.579388e+01
## mean............ 1.561801e+01
## variance........ 3.113896e+01
## var 2:
## best............ 8.068548e+02
## mean............ 8.141974e+02
## variance........ 6.784355e+02
## var 3:
## best............ 3.115044e+02
## mean............ 3.153912e+02
## variance........ 4.152574e+02
## var 4:
## best............ 8.926560e+02
## mean............ 8.295218e+02
## variance........ 2.315411e+04
## var 5:
## best............ 8.872365e+02
## mean............ 8.659688e+02
## variance........ 8.260440e+03
## var 6:
## best............ 3.549239e+02
## mean............ 4.252438e+02
## variance........ 4.923746e+03
## var 7:
## best............ 3.313543e+01
## mean............ 1.724074e+02
## variance........ 8.139028e+04
##
## GENERATION: 4
## Lexical Fit..... 3.173108e-01 3.173108e-01 3.300000e-01 3.701437e-01 3.701437e-01 4.120290e-01 5.271645e-01 5.271645e-01 5.827818e-01 8.737061e-01 9.720000e-01 9.980000e-01 1.000000e+00 1.000000e+00
## #unique......... 11, #Total UniqueCount: 66
## var 1:
## best............ 1.579388e+01
## mean............ 2.808238e+01
## variance........ 1.915818e+03
## var 2:
## best............ 8.068548e+02
## mean............ 7.575442e+02
## variance........ 2.100367e+04
## var 3:
## best............ 3.115044e+02
## mean............ 3.092786e+02
## variance........ 2.548205e+03
## var 4:
## best............ 8.926560e+02
## mean............ 8.954764e+02
## variance........ 7.263628e+01
## var 5:
## best............ 8.872365e+02
## mean............ 8.877601e+02
## variance........ 8.330646e+02
## var 6:
## best............ 3.549239e+02
## mean............ 4.347976e+02
## variance........ 9.407492e+03
## var 7:
## best............ 3.313543e+01
## mean............ 7.299991e+01
## variance........ 1.512679e+04
##
## GENERATION: 5
## Lexical Fit..... 3.173108e-01 3.173108e-01 3.300000e-01 3.701437e-01 3.701437e-01 4.120290e-01 5.271645e-01 5.271645e-01 5.827818e-01 8.737061e-01 9.720000e-01 9.980000e-01 1.000000e+00 1.000000e+00
## #unique......... 10, #Total UniqueCount: 76
## var 1:
## best............ 1.579388e+01
## mean............ 1.424302e+01
## variance........ 8.437568e+00
## var 2:
## best............ 8.068548e+02
## mean............ 8.033015e+02
## variance........ 1.614088e+03
## var 3:
## best............ 3.115044e+02
## mean............ 3.283206e+02
## variance........ 3.645326e+03
## var 4:
## best............ 8.926560e+02
## mean............ 8.931657e+02
## variance........ 2.775267e+00
## var 5:
## best............ 8.872365e+02
## mean............ 8.924514e+02
## variance........ 1.589678e+02
## var 6:
## best............ 3.549239e+02
## mean............ 4.157515e+02
## variance........ 2.112952e+04
## var 7:
## best............ 3.313543e+01
## mean............ 4.373805e+01
## variance........ 1.662836e+03
##
## GENERATION: 6
## Lexical Fit..... 3.173108e-01 3.173108e-01 3.300000e-01 3.701437e-01 3.701437e-01 4.120290e-01 5.271645e-01 5.271645e-01 5.827818e-01 8.737061e-01 9.720000e-01 9.980000e-01 1.000000e+00 1.000000e+00
## #unique......... 8, #Total UniqueCount: 84
## var 1:
## best............ 1.579388e+01
## mean............ 4.052015e+01
## variance........ 6.770649e+03
## var 2:
## best............ 8.068548e+02
## mean............ 8.029304e+02
## variance........ 1.123459e+03
## var 3:
## best............ 3.115044e+02
## mean............ 3.241346e+02
## variance........ 3.560530e+03
## var 4:
## best............ 8.926560e+02
## mean............ 8.707818e+02
## variance........ 4.661095e+03
## var 5:
## best............ 8.872365e+02
## mean............ 8.882447e+02
## variance........ 8.386752e+01
## var 6:
## best............ 3.549239e+02
## mean............ 3.725770e+02
## variance........ 5.414634e+03
## var 7:
## best............ 3.313543e+01
## mean............ 5.212469e+01
## variance........ 3.797123e+03
##
## GENERATION: 7
## Lexical Fit..... 3.173108e-01 3.173108e-01 3.300000e-01 3.701437e-01 3.701437e-01 4.120290e-01 5.271645e-01 5.271645e-01 5.827818e-01 8.737061e-01 9.720000e-01 9.980000e-01 1.000000e+00 1.000000e+00
## #unique......... 6, #Total UniqueCount: 90
## var 1:
## best............ 1.579388e+01
## mean............ 3.926465e+01
## variance........ 6.727968e+03
## var 2:
## best............ 8.068548e+02
## mean............ 8.027507e+02
## variance........ 1.161955e+02
## var 3:
## best............ 3.115044e+02
## mean............ 3.113771e+02
## variance........ 2.862367e+02
## var 4:
## best............ 8.926560e+02
## mean............ 8.934560e+02
## variance........ 2.492677e+01
## var 5:
## best............ 8.872365e+02
## mean............ 8.749845e+02
## variance........ 2.172842e+03
## var 6:
## best............ 3.549239e+02
## mean............ 3.508345e+02
## variance........ 2.522213e+02
## var 7:
## best............ 3.313543e+01
## mean............ 5.940284e+01
## variance........ 1.137567e+04
##
## GENERATION: 8
## Lexical Fit..... 3.173108e-01 3.173108e-01 3.300000e-01 3.701437e-01 3.701437e-01 4.120290e-01 5.271645e-01 5.271645e-01 5.827818e-01 8.737061e-01 9.720000e-01 9.980000e-01 1.000000e+00 1.000000e+00
## #unique......... 7, #Total UniqueCount: 97
## var 1:
## best............ 1.579388e+01
## mean............ 2.439233e+01
## variance........ 7.679483e+02
## var 2:
## best............ 8.068548e+02
## mean............ 7.793968e+02
## variance........ 1.651973e+04
## var 3:
## best............ 3.115044e+02
## mean............ 3.011862e+02
## variance........ 3.000053e+03
## var 4:
## best............ 8.926560e+02
## mean............ 8.558825e+02
## variance........ 1.054064e+04
## var 5:
## best............ 8.872365e+02
## mean............ 8.647865e+02
## variance........ 9.941736e+03
## var 6:
## best............ 3.549239e+02
## mean............ 3.522310e+02
## variance........ 5.081417e+02
## var 7:
## best............ 3.313543e+01
## mean............ 1.084554e+02
## variance........ 3.938919e+04
##
## 'wait.generations' limit reached.
## No significant improvement in 4 generations.
##
## Solution Lexical Fitness Value:
## 3.173108e-01 3.173108e-01 3.300000e-01 3.701437e-01 3.701437e-01 4.120290e-01 5.271645e-01 5.271645e-01 5.827818e-01 8.737061e-01 9.720000e-01 9.980000e-01 1.000000e+00 1.000000e+00
##
## Parameters at the Solution:
##
## X[ 1] : 1.579388e+01
## X[ 2] : 8.068548e+02
## X[ 3] : 3.115044e+02
## X[ 4] : 8.926560e+02
## X[ 5] : 8.872365e+02
## X[ 6] : 3.549239e+02
## X[ 7] : 3.313543e+01
##
## Solution Found Generation 3
## Number of Generations Run 8
##
## Fri Apr 22 18:55:59 2022
## Total run time : 0 hours 1 minutes and 24 seconds
mout <- Match(Y = data_omit$agree.evoting, Tr = data_omit$EV, X = cbind(data_omit$age.group, data_omit$educ, data_omit$male, data_omit$tech, data_omit$pol.info, data_omit$white.collar, data_omit$not.full.time), Weight.matrix = genout)
mb_gm <- MatchBalance(EV ~ age.group + educ + male + tech + pol.info + white.collar + not.full.time, data = data_omit, match.out = mout)
##
## ***** (V1) age.group *****
## Before Matching After Matching
## mean treatment........ 2.4673 2.4673
## mean control.......... 2.446 2.4368
## std mean diff......... 1.5948 2.2766
##
## mean raw eQQ diff..... 0.063465 0.074074
## med raw eQQ diff..... 0 0
## max raw eQQ diff..... 1 1
##
## mean eCDF diff........ 0.012531 0.014815
## med eCDF diff........ 0.017628 0.019157
## max eCDF diff........ 0.024371 0.024904
##
## var ratio (Tr/Co)..... 1.0361 1.0932
## T-test p-value........ 0.76583 0.41203
## KS Bootstrap p-value.. 0.62 0.314
## KS Naive p-value...... 0.98724 0.71643
## KS Statistic.......... 0.024371 0.024904
##
##
## ***** (V2) educ *****
## Before Matching After Matching
## mean treatment........ 4.7615 4.7615
## mean control.......... 4.0909 4.7518
## std mean diff......... 29.258 0.42256
##
## mean raw eQQ diff..... 0.66724 0.03576
## med raw eQQ diff..... 1 0
## max raw eQQ diff..... 2 2
##
## mean eCDF diff........ 0.075703 0.0039911
## med eCDF diff........ 0.08996 0.0047893
## max eCDF diff........ 0.14546 0.0063857
##
## var ratio (Tr/Co)..... 1.3492 1.0329
## T-test p-value........ 5.3884e-09 0.58278
## KS Bootstrap p-value.. < 2.22e-16 0.998
## KS Naive p-value...... 1.047e-06 1
## KS Statistic.......... 0.14546 0.0063857
##
##
## ***** (V3) male *****
## Before Matching After Matching
## mean treatment........ 0.49879 0.49879
## mean control.......... 0.49571 0.49637
## std mean diff......... 0.61513 0.48397
##
## mean raw eQQ diff..... 0.0017153 0.0012771
## med raw eQQ diff..... 0 0
## max raw eQQ diff..... 1 1
##
## mean eCDF diff........ 0.0015388 0.00063857
## med eCDF diff........ 0.0015388 0.00063857
## max eCDF diff........ 0.0030775 0.0012771
##
## var ratio (Tr/Co)..... 0.99956 1
## T-test p-value........ 0.90949 0.52716
##
##
## ***** (V4) tech *****
## Before Matching After Matching
## mean treatment........ 4.1949 4.1949
## mean control.......... 3.916 4.1931
## std mean diff......... 17.053 0.11101
##
## mean raw eQQ diff..... 0.27616 0.018519
## med raw eQQ diff..... 0 0
## max raw eQQ diff..... 1 1
##
## mean eCDF diff........ 0.046494 0.0030864
## med eCDF diff........ 0.044928 0.0031928
## max eCDF diff........ 0.1073 0.0083014
##
## var ratio (Tr/Co)..... 1.0308 1.0325
## T-test p-value........ 0.0015052 0.87371
## KS Bootstrap p-value.. < 2.22e-16 0.972
## KS Naive p-value...... 0.00076398 1
## KS Statistic.......... 0.1073 0.0083014
##
##
## ***** (V5) pol.info *****
## Before Matching After Matching
## mean treatment........ 1.4685 1.4685
## mean control.......... 1.3053 1.4685
## std mean diff......... 20.407 0
##
## mean raw eQQ diff..... 0.16295 0
## med raw eQQ diff..... 0 0
## max raw eQQ diff..... 1 0
##
## mean eCDF diff........ 0.040801 0
## med eCDF diff........ 0.035003 0
## max eCDF diff........ 0.0932 0
##
## var ratio (Tr/Co)..... 1.4448 1
## T-test p-value........ 3.2741e-05 1
## KS Bootstrap p-value.. < 2.22e-16 1
## KS Naive p-value...... 0.0052778 1
## KS Statistic.......... 0.0932 0
##
##
## ***** (V6) white.collar *****
## Before Matching After Matching
## mean treatment........ 0.30508 0.30508
## mean control.......... 0.2813 0.30387
## std mean diff......... 5.1617 0.26277
##
## mean raw eQQ diff..... 0.024014 0.00063857
## med raw eQQ diff..... 0 0
## max raw eQQ diff..... 1 1
##
## mean eCDF diff........ 0.011891 0.00031928
## med eCDF diff........ 0.011891 0.00031928
## max eCDF diff........ 0.023781 0.00063857
##
## var ratio (Tr/Co)..... 1.0481 1.0022
## T-test p-value........ 0.33355 0.31731
##
##
## ***** (V7) not.full.time *****
## Before Matching After Matching
## mean treatment........ 0.27603 0.27603
## mean control.......... 0.33448 0.26755
## std mean diff......... -13.067 1.8946
##
## mean raw eQQ diff..... 0.058319 0.0083014
## med raw eQQ diff..... 0 0
## max raw eQQ diff..... 1 1
##
## mean eCDF diff........ 0.029224 0.0041507
## med eCDF diff........ 0.029224 0.0041507
## max eCDF diff........ 0.058448 0.0083014
##
## var ratio (Tr/Co)..... 0.89728 1.0197
## T-test p-value........ 0.019525 0.37014
##
##
## Before Matching Minimum p.value: < 2.22e-16
## Variable Name(s): educ tech pol.info Number(s): 2 4 5
##
## After Matching Minimum p.value: 0.314
## Variable Name(s): age.group Number(s): 1
summary(mout)
##
## Estimate... 0.18634
## AI SE...... 0.029299
## T-stat..... 6.3598
## p.val...... 2.0205e-10
##
## Original number of observations.............. 1409
## Original number of treated obs............... 826
## Matched number of observations............... 826
## Matched number of observations (unweighted). 1566
# EXACT MATCHING ON BINARY VARIABLE
set.seed(34664)
# we impose EXACT Matching on Male, White Collar and Not Full Time!
genout1 <- GenMatch(Tr = data_omit$EV, X = cbind(data_omit$age.group, data_omit$educ, data_omit$male, data_omit$tech, data_omit$pol.info, data_omit$white.collar, data_omit$not.full.time), exact = c(FALSE, FALSE, TRUE, FALSE, FALSE, TRUE, TRUE), pop.size = 20, nboots = 500) # increase the number of bootstraps to increase the quality of matching
##
##
## Fri Apr 22 18:56:02 2022
## Domains:
## 0.000000e+00 <= X1 <= 1.000000e+03
## 0.000000e+00 <= X2 <= 1.000000e+03
## 0.000000e+00 <= X3 <= 1.000000e+03
## 0.000000e+00 <= X4 <= 1.000000e+03
## 0.000000e+00 <= X5 <= 1.000000e+03
## 0.000000e+00 <= X6 <= 1.000000e+03
## 0.000000e+00 <= X7 <= 1.000000e+03
##
## Data Type: Floating Point
## Operators (code number, name, population)
## (1) Cloning........................... 5
## (2) Uniform Mutation.................. 2
## (3) Boundary Mutation................. 2
## (4) Non-Uniform Mutation.............. 2
## (5) Polytope Crossover................ 2
## (6) Simple Crossover.................. 2
## (7) Whole Non-Uniform Mutation........ 2
## (8) Heuristic Crossover............... 2
## (9) Local-Minimum Crossover........... 0
##
## SOFT Maximum Number of Generations: 100
## Maximum Nonchanging Generations: 4
## Population size : 20
## Convergence Tolerance: 1.000000e-03
##
## Not Using the BFGS Derivative Based Optimizer on the Best Individual Each Generation.
## Not Checking Gradients before Stopping.
## Using Out of Bounds Individuals.
##
## Maximization Problem.
## GENERATION: 0 (initializing the population)
## Lexical Fit..... 4.836045e-03 8.307797e-02 1.485014e-01 3.100000e-01 6.960000e-01 7.570985e-01 9.000000e-01 9.860000e-01 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00
## #unique......... 20, #Total UniqueCount: 20
## var 1:
## best............ 1.379989e+02
## mean............ 3.339641e+02
## variance........ 6.450582e+04
## var 2:
## best............ 1.428679e+02
## mean............ 4.941728e+02
## variance........ 8.645013e+04
## var 3:
## best............ 6.925637e+02
## mean............ 4.384428e+02
## variance........ 1.011329e+05
## var 4:
## best............ 5.295281e+01
## mean............ 4.598639e+02
## variance........ 9.014094e+04
## var 5:
## best............ 9.974096e+02
## mean............ 5.452365e+02
## variance........ 8.401963e+04
## var 6:
## best............ 9.609640e+01
## mean............ 3.831739e+02
## variance........ 9.354026e+04
## var 7:
## best............ 7.630937e+01
## mean............ 3.950401e+02
## variance........ 7.549191e+04
##
## GENERATION: 1
## Lexical Fit..... 1.186951e-02 8.307797e-02 3.685363e-01 3.783936e-01 5.740000e-01 6.940000e-01 8.740000e-01 9.920000e-01 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00
## #unique......... 13, #Total UniqueCount: 33
## var 1:
## best............ 1.286089e+02
## mean............ 1.888537e+02
## variance........ 2.376633e+04
## var 2:
## best............ 2.018664e+02
## mean............ 3.329961e+02
## variance........ 6.466280e+04
## var 3:
## best............ 3.336807e+02
## mean............ 6.226979e+02
## variance........ 5.277562e+04
## var 4:
## best............ 1.094720e+02
## mean............ 2.729393e+02
## variance........ 6.349704e+04
## var 5:
## best............ 9.403113e+02
## mean............ 7.753558e+02
## variance........ 3.775796e+04
## var 6:
## best............ 1.632464e+02
## mean............ 3.562311e+02
## variance........ 7.861835e+04
## var 7:
## best............ 1.676028e+02
## mean............ 3.485271e+02
## variance........ 1.137334e+05
##
## GENERATION: 2
## Lexical Fit..... 2.181549e-02 8.307797e-02 1.984772e-01 5.220000e-01 5.669920e-01 8.260000e-01 9.840000e-01 9.960000e-01 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00
## #unique......... 12, #Total UniqueCount: 45
## var 1:
## best............ 1.286089e+02
## mean............ 1.297390e+02
## variance........ 1.900780e+02
## var 2:
## best............ 2.018664e+02
## mean............ 1.933430e+02
## variance........ 1.952607e+04
## var 3:
## best............ 3.336807e+02
## mean............ 5.529596e+02
## variance........ 3.499129e+04
## var 4:
## best............ 1.798559e+02
## mean............ 8.576332e+01
## variance........ 2.570888e+03
## var 5:
## best............ 9.403113e+02
## mean............ 9.313420e+02
## variance........ 6.893506e+03
## var 6:
## best............ 1.632464e+02
## mean............ 1.322592e+02
## variance........ 6.042703e+03
## var 7:
## best............ 1.676028e+02
## mean............ 1.635550e+02
## variance........ 2.327517e+04
##
## GENERATION: 3
## Lexical Fit..... 4.698195e-02 8.307797e-02 1.469536e-01 4.833818e-01 6.000000e-01 8.340000e-01 9.900000e-01 9.960000e-01 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00
## #unique......... 10, #Total UniqueCount: 55
## var 1:
## best............ 1.012600e+02
## mean............ 1.219728e+02
## variance........ 4.576029e+02
## var 2:
## best............ 1.835195e+02
## mean............ 2.166273e+02
## variance........ 1.253460e+04
## var 3:
## best............ 3.290847e+02
## mean............ 3.661672e+02
## variance........ 5.650683e+03
## var 4:
## best............ 1.026700e+02
## mean............ 1.543721e+02
## variance........ 1.080190e+04
## var 5:
## best............ 9.463827e+02
## mean............ 9.337782e+02
## variance........ 1.626057e+03
## var 6:
## best............ 1.436520e+02
## mean............ 1.396995e+02
## variance........ 1.556008e+03
## var 7:
## best............ 2.282243e+02
## mean............ 2.080455e+02
## variance........ 7.776701e+03
##
## GENERATION: 4
## Lexical Fit..... 4.698195e-02 8.307797e-02 1.469536e-01 4.833818e-01 6.000000e-01 8.340000e-01 9.900000e-01 9.960000e-01 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00
## #unique......... 10, #Total UniqueCount: 65
## var 1:
## best............ 1.012600e+02
## mean............ 1.592652e+02
## variance........ 1.732290e+04
## var 2:
## best............ 1.835195e+02
## mean............ 2.239877e+02
## variance........ 1.116085e+04
## var 3:
## best............ 3.290847e+02
## mean............ 3.269583e+02
## variance........ 9.356424e+03
## var 4:
## best............ 1.026700e+02
## mean............ 1.697697e+02
## variance........ 1.775181e+04
## var 5:
## best............ 9.463827e+02
## mean............ 9.243144e+02
## variance........ 8.381441e+03
## var 6:
## best............ 1.436520e+02
## mean............ 1.694995e+02
## variance........ 8.824852e+03
## var 7:
## best............ 2.282243e+02
## mean............ 2.149256e+02
## variance........ 3.378055e+03
##
## GENERATION: 5
## Lexical Fit..... 4.698195e-02 8.307797e-02 1.469536e-01 4.833818e-01 6.000000e-01 8.340000e-01 9.900000e-01 9.960000e-01 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00
## #unique......... 6, #Total UniqueCount: 71
## var 1:
## best............ 1.012600e+02
## mean............ 1.037672e+02
## variance........ 2.212374e+02
## var 2:
## best............ 1.835195e+02
## mean............ 2.087744e+02
## variance........ 1.836008e+04
## var 3:
## best............ 3.290847e+02
## mean............ 3.182633e+02
## variance........ 4.147378e+03
## var 4:
## best............ 1.026700e+02
## mean............ 1.421198e+02
## variance........ 8.688824e+03
## var 5:
## best............ 9.463827e+02
## mean............ 9.468128e+02
## variance........ 1.546689e+01
## var 6:
## best............ 1.436520e+02
## mean............ 1.825567e+02
## variance........ 8.310971e+03
## var 7:
## best............ 2.282243e+02
## mean............ 2.247847e+02
## variance........ 1.127122e+03
##
## GENERATION: 6
## Lexical Fit..... 4.698195e-02 8.307797e-02 1.469536e-01 4.833818e-01 6.000000e-01 8.340000e-01 9.900000e-01 9.960000e-01 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00
## #unique......... 7, #Total UniqueCount: 78
## var 1:
## best............ 1.012600e+02
## mean............ 1.125381e+02
## variance........ 1.613987e+03
## var 2:
## best............ 1.835195e+02
## mean............ 1.893524e+02
## variance........ 7.873928e+02
## var 3:
## best............ 3.290847e+02
## mean............ 3.158317e+02
## variance........ 1.664312e+03
## var 4:
## best............ 1.026700e+02
## mean............ 1.170266e+02
## variance........ 1.910750e+03
## var 5:
## best............ 9.463827e+02
## mean............ 9.287002e+02
## variance........ 5.436280e+03
## var 6:
## best............ 1.436520e+02
## mean............ 1.356895e+02
## variance........ 3.308251e+02
## var 7:
## best............ 2.282243e+02
## mean............ 2.301199e+02
## variance........ 9.634211e+02
##
## GENERATION: 7
## Lexical Fit..... 4.698195e-02 8.307797e-02 1.469536e-01 4.833818e-01 6.020000e-01 8.580000e-01 9.900000e-01 9.960000e-01 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00
## #unique......... 8, #Total UniqueCount: 86
## var 1:
## best............ 1.012600e+02
## mean............ 1.753295e+02
## variance........ 3.864865e+04
## var 2:
## best............ 1.835195e+02
## mean............ 2.336020e+02
## variance........ 1.309365e+04
## var 3:
## best............ 3.290847e+02
## mean............ 3.495206e+02
## variance........ 1.052046e+04
## var 4:
## best............ 1.026700e+02
## mean............ 1.737563e+02
## variance........ 4.039109e+04
## var 5:
## best............ 9.463827e+02
## mean............ 9.350146e+02
## variance........ 1.379404e+03
## var 6:
## best............ 1.436520e+02
## mean............ 1.622761e+02
## variance........ 9.767151e+03
## var 7:
## best............ 1.762649e+02
## mean............ 2.505425e+02
## variance........ 1.009147e+04
##
## GENERATION: 8
## Lexical Fit..... 4.698195e-02 8.307797e-02 1.469536e-01 4.833818e-01 6.020000e-01 8.580000e-01 9.900000e-01 9.960000e-01 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00
## #unique......... 11, #Total UniqueCount: 97
## var 1:
## best............ 1.012600e+02
## mean............ 1.142494e+02
## variance........ 1.264141e+03
## var 2:
## best............ 1.835195e+02
## mean............ 2.000427e+02
## variance........ 7.560549e+03
## var 3:
## best............ 3.290847e+02
## mean............ 3.204486e+02
## variance........ 9.064791e+03
## var 4:
## best............ 1.026700e+02
## mean............ 1.420088e+02
## variance........ 9.640964e+03
## var 5:
## best............ 9.463827e+02
## mean............ 9.238820e+02
## variance........ 3.465513e+03
## var 6:
## best............ 1.436520e+02
## mean............ 1.464330e+02
## variance........ 2.790972e+02
## var 7:
## best............ 1.762649e+02
## mean............ 2.601278e+02
## variance........ 2.059823e+04
##
## GENERATION: 9
## Lexical Fit..... 4.698195e-02 8.307797e-02 1.469536e-01 4.833818e-01 6.020000e-01 8.580000e-01 9.900000e-01 9.960000e-01 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00
## #unique......... 8, #Total UniqueCount: 105
## var 1:
## best............ 1.012600e+02
## mean............ 9.926508e+01
## variance........ 7.561707e+01
## var 2:
## best............ 1.835195e+02
## mean............ 1.834320e+02
## variance........ 2.802076e+01
## var 3:
## best............ 3.290847e+02
## mean............ 3.331346e+02
## variance........ 1.117152e+03
## var 4:
## best............ 1.026700e+02
## mean............ 1.018587e+02
## variance........ 1.150102e+01
## var 5:
## best............ 9.463827e+02
## mean............ 9.247192e+02
## variance........ 4.507005e+03
## var 6:
## best............ 1.436520e+02
## mean............ 1.741293e+02
## variance........ 1.391250e+04
## var 7:
## best............ 1.762649e+02
## mean............ 2.123327e+02
## variance........ 5.239741e+03
##
## GENERATION: 10
## Lexical Fit..... 4.698195e-02 8.307797e-02 1.469536e-01 4.833818e-01 6.020000e-01 8.580000e-01 9.900000e-01 9.960000e-01 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00
## #unique......... 6, #Total UniqueCount: 111
## var 1:
## best............ 1.012600e+02
## mean............ 1.468670e+02
## variance........ 1.302637e+04
## var 2:
## best............ 1.835195e+02
## mean............ 2.474826e+02
## variance........ 3.651518e+04
## var 3:
## best............ 3.290847e+02
## mean............ 3.483456e+02
## variance........ 4.147389e+03
## var 4:
## best............ 1.026700e+02
## mean............ 1.157850e+02
## variance........ 1.716581e+03
## var 5:
## best............ 9.463827e+02
## mean............ 9.261408e+02
## variance........ 3.734761e+03
## var 6:
## best............ 1.436520e+02
## mean............ 2.048016e+02
## variance........ 2.847460e+04
## var 7:
## best............ 1.762649e+02
## mean............ 2.108481e+02
## variance........ 7.565286e+03
##
## GENERATION: 11
## Lexical Fit..... 4.698195e-02 8.307797e-02 1.469536e-01 4.833818e-01 6.020000e-01 8.580000e-01 9.900000e-01 9.960000e-01 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00
## #unique......... 9, #Total UniqueCount: 120
## var 1:
## best............ 1.012600e+02
## mean............ 1.330931e+02
## variance........ 9.695272e+03
## var 2:
## best............ 1.835195e+02
## mean............ 1.818152e+02
## variance........ 1.843684e+02
## var 3:
## best............ 3.290847e+02
## mean............ 3.541779e+02
## variance........ 2.037047e+04
## var 4:
## best............ 1.026700e+02
## mean............ 1.057149e+02
## variance........ 8.397803e+01
## var 5:
## best............ 9.463827e+02
## mean............ 9.128774e+02
## variance........ 1.886419e+04
## var 6:
## best............ 1.436520e+02
## mean............ 1.665790e+02
## variance........ 1.338681e+04
## var 7:
## best............ 1.762649e+02
## mean............ 1.744509e+02
## variance........ 5.130763e+01
##
## GENERATION: 12
## Lexical Fit..... 4.698195e-02 8.307797e-02 1.469536e-01 4.833818e-01 6.020000e-01 8.580000e-01 9.900000e-01 9.960000e-01 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00
## #unique......... 6, #Total UniqueCount: 126
## var 1:
## best............ 1.012600e+02
## mean............ 1.130725e+02
## variance........ 2.763320e+03
## var 2:
## best............ 1.835195e+02
## mean............ 1.822327e+02
## variance........ 2.699412e+02
## var 3:
## best............ 3.290847e+02
## mean............ 3.385277e+02
## variance........ 2.233328e+03
## var 4:
## best............ 1.026700e+02
## mean............ 1.613277e+02
## variance........ 2.961529e+04
## var 5:
## best............ 9.463827e+02
## mean............ 9.234178e+02
## variance........ 5.025230e+03
## var 6:
## best............ 1.436520e+02
## mean............ 1.455169e+02
## variance........ 7.204419e+01
## var 7:
## best............ 1.762649e+02
## mean............ 1.861917e+02
## variance........ 1.929544e+03
##
## 'wait.generations' limit reached.
## No significant improvement in 4 generations.
##
## Solution Lexical Fitness Value:
## 4.698195e-02 8.307797e-02 1.469536e-01 4.833818e-01 6.020000e-01 8.580000e-01 9.900000e-01 9.960000e-01 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00
##
## Parameters at the Solution:
##
## X[ 1] : 1.012600e+02
## X[ 2] : 1.835195e+02
## X[ 3] : 3.290847e+02
## X[ 4] : 1.026700e+02
## X[ 5] : 9.463827e+02
## X[ 6] : 1.436520e+02
## X[ 7] : 1.762649e+02
##
## Solution Found Generation 7
## Number of Generations Run 12
##
## Fri Apr 22 18:57:49 2022
## Total run time : 0 hours 1 minutes and 47 seconds
mout1 <- Match(Y = data_omit$agree.evoting, Tr = data_omit$EV, X = cbind(data_omit$age.group, data_omit$educ, data_omit$male, data_omit$tech, data_omit$pol.info, data_omit$white.collar, data_omit$not.full.time), Weight.matrix = genout1)
mb_gm1 <- MatchBalance(EV ~ age.group + educ + male + tech + pol.info + white.collar + not.full.time, data = data_omit, match.out = mout1)
##
## ***** (V1) age.group *****
## Before Matching After Matching
## mean treatment........ 2.4673 2.4673
## mean control.......... 2.446 2.458
## std mean diff......... 1.5948 0.69354
##
## mean raw eQQ diff..... 0.063465 0.032692
## med raw eQQ diff..... 0 0
## max raw eQQ diff..... 1 1
##
## mean eCDF diff........ 0.012531 0.0065385
## med eCDF diff........ 0.017628 0.0083333
## max eCDF diff........ 0.024371 0.010897
##
## var ratio (Tr/Co)..... 1.0361 1.073
## T-test p-value........ 0.76583 0.63388
## KS Bootstrap p-value.. 0.66 0.844
## KS Naive p-value...... 0.98724 0.99999
## KS Statistic.......... 0.024371 0.010897
##
##
## ***** (V2) educ *****
## Before Matching After Matching
## mean treatment........ 4.7615 4.7615
## mean control.......... 4.0909 4.7378
## std mean diff......... 29.258 1.0353
##
## mean raw eQQ diff..... 0.66724 0.058974
## med raw eQQ diff..... 1 0
## max raw eQQ diff..... 2 2
##
## mean eCDF diff........ 0.075703 0.0061699
## med eCDF diff........ 0.08996 0.0044872
## max eCDF diff........ 0.14546 0.015385
##
## var ratio (Tr/Co)..... 1.3492 1.0871
## T-test p-value........ 5.3884e-09 0.39236
## KS Bootstrap p-value.. < 2.22e-16 0.774
## KS Naive p-value...... 1.047e-06 0.99269
## KS Statistic.......... 0.14546 0.015385
##
##
## ***** (V3) male *****
## Before Matching After Matching
## mean treatment........ 0.49879 0.49879
## mean control.......... 0.49571 0.49879
## std mean diff......... 0.61513 0
##
## mean raw eQQ diff..... 0.0017153 0
## med raw eQQ diff..... 0 0
## max raw eQQ diff..... 1 0
##
## mean eCDF diff........ 0.0015388 0
## med eCDF diff........ 0.0015388 0
## max eCDF diff........ 0.0030775 0
##
## var ratio (Tr/Co)..... 0.99956 1
## T-test p-value........ 0.90949 1
##
##
## ***** (V4) tech *****
## Before Matching After Matching
## mean treatment........ 4.1949 4.1949
## mean control.......... 3.916 4.2324
## std mean diff......... 17.053 -2.2943
##
## mean raw eQQ diff..... 0.27616 0.016026
## med raw eQQ diff..... 0 0
## max raw eQQ diff..... 1 1
##
## mean eCDF diff........ 0.046494 0.0026709
## med eCDF diff........ 0.044928 0.0019231
## max eCDF diff........ 0.1073 0.0064103
##
## var ratio (Tr/Co)..... 1.0308 1.0617
## T-test p-value........ 0.0015052 0.068985
## KS Bootstrap p-value.. < 2.22e-16 0.99
## KS Naive p-value...... 0.00076398 1
## KS Statistic.......... 0.1073 0.0064103
##
##
## ***** (V5) pol.info *****
## Before Matching After Matching
## mean treatment........ 1.4685 1.4685
## mean control.......... 1.3053 1.4673
## std mean diff......... 20.407 0.15138
##
## mean raw eQQ diff..... 0.16295 0.00064103
## med raw eQQ diff..... 0 0
## max raw eQQ diff..... 1 1
##
## mean eCDF diff........ 0.040801 0.00016026
## med eCDF diff........ 0.035003 0
## max eCDF diff........ 0.0932 0.00064103
##
## var ratio (Tr/Co)..... 1.4448 1.0078
## T-test p-value........ 3.2741e-05 0.31731
## KS Bootstrap p-value.. < 2.22e-16 1
## KS Naive p-value...... 0.0052778 1
## KS Statistic.......... 0.0932 0.00064103
##
##
## ***** (V6) white.collar *****
## Before Matching After Matching
## mean treatment........ 0.30508 0.30508
## mean control.......... 0.2813 0.29903
## std mean diff......... 5.1617 1.3139
##
## mean raw eQQ diff..... 0.024014 0.0032051
## med raw eQQ diff..... 0 0
## max raw eQQ diff..... 1 1
##
## mean eCDF diff........ 0.011891 0.0016026
## med eCDF diff........ 0.011891 0.0016026
## max eCDF diff........ 0.023781 0.0032051
##
## var ratio (Tr/Co)..... 1.0481 1.0114
## T-test p-value........ 0.33355 0.02517
##
##
## ***** (V7) not.full.time *****
## Before Matching After Matching
## mean treatment........ 0.27603 0.27603
## mean control.......... 0.33448 0.2724
## std mean diff......... -13.067 0.81197
##
## mean raw eQQ diff..... 0.058319 0.0019231
## med raw eQQ diff..... 0 0
## max raw eQQ diff..... 1 1
##
## mean eCDF diff........ 0.029224 0.00096154
## med eCDF diff........ 0.029224 0.00096154
## max eCDF diff........ 0.058448 0.0019231
##
## var ratio (Tr/Co)..... 0.89728 1.0083
## T-test p-value........ 0.019525 0.083078
##
##
## Before Matching Minimum p.value: < 2.22e-16
## Variable Name(s): educ tech pol.info Number(s): 2 4 5
##
## After Matching Minimum p.value: 0.02517
## Variable Name(s): white.collar Number(s): 6
summary(mout1)
##
## Estimate... 0.1928
## AI SE...... 0.028555
## T-stat..... 6.7517
## p.val...... 1.461e-11
##
## Original number of observations.............. 1409
## Original number of treated obs............... 826
## Matched number of observations............... 826
## Matched number of observations (unweighted). 1560