Spielplatz/test_direct.R
In SoerenPannier/emdi: Estimating and Mapping Disaggregated Indicators
## New Function direct
library(laeken)
library(emdi)
data(eusilc)
# 1. with weights and sampling domains (naive bootstrap)
direct_all_naive <- direct(y = "eqIncome",
smp_data = eusilc,
smp_domains = "db040",
weights = "rb050",
# without weights
# with weights
#threshold=10859.24,
threshold = function(y,weights){0.6 * weightedMedian(x = y, weights = weights)},
var=TRUE,
#bootType = "naive",
boot_type = "naive",
X = NULL,
totals = NULL,
B=50,
seed=123,
na.rm=TRUE)
# laeken indicators that fit to above setting
arpr_all <- arpr(eusilc$eqIncome, breakdown = eusilc$db040, weights=eusilc$rb050)
arpr_all_naive <- variance("eqIncome",
weights="rb050",
data=eusilc,
breakdown = "db040",
indicator=arpr_all,
R=50,
bootType="naive",
X = NULL,
seed=123)
# Check correspondence: difference in variance can occur due to different poverty
# lines
all.equal(arpr_all$valueByStratum, direct_all_naive$ind[, c("Domain","Head_Count")])
all.equal(arpr_all_naive$varByStratum, direct_all_naive$MSE[, c("Domain","Head_Count")])
all.equal(arpr_all_naive$varByStratum$var/10000, direct_all_naive$MSE[, c("Domain","Head_Count")][,2])
gini_all <- gini(eusilc$eqIncome, breakdown = eusilc$db040, weights=eusilc$rb050)
gini_all_naive <- variance("eqIncome",
weights="rb050",
data=eusilc,
breakdown = "db040",
bootType="naive",
indicator=gini_all,
R=50,
seed=123)
# Check correspondence
all.equal(gini_all$valueByStratum, direct_all_naive$ind[, c("Domain","Gini")])
all.equal(gini_all_naive$varByStratum, direct_all_naive$MSE[, c("Domain","Gini")])
qsr_all <- qsr(eusilc$eqIncome,breakdown = eusilc$db040, weights=eusilc$rb050)
qsr_all_naive <- variance("eqIncome",
weights="rb050",
data=eusilc,
breakdown = "db040",
bootType="naive",
X = NULL,
indicator=qsr_all,
R=50,
seed=123)
# Check correspondence
all.equal(qsr_all$valueByStratum, direct_all_naive$ind[, c("Domain","Quintile_Share")])
all.equal(qsr_all_naive$varByStratum, direct_all_naive$MSE[, c("Domain","Quintile_Share")])
# 2. with weights and sampling domains (calibrate bootstrap)
direct_all_cali <- direct(y="eqIncome",
smp_data=eusilc,
smp_domains="db040",
weights="rb050",
# without weights
#threshold = 10848.8,
# with weights
threshold=10859.24,
var=TRUE,
bootType = "calibrate",
X = as.matrix(eusilc$age),
totals = NULL,
B=50,
seed=123,
na.rm=TRUE)
direct_all_cali <- direct(y="eqIncome",
smp_data=eusilc,
smp_domains="db040",
weights="rb050",
# without weights
#threshold = 10848.8,
# with weights
threshold=10859.24,
var=TRUE,
boot_type = "calibrate",
X_calib = as.matrix(eusilc$age),
totals = NULL,
B=50,
seed=123,
na.rm=TRUE)
# laeken indicators that fit to above setting
arpr_all_cali <- variance("eqIncome",
weights="rb050",
data=eusilc,
breakdown = "db040",
indicator=arpr_all,
R=50,
bootType="calibrate",
X = as.matrix(eusilc$age),
seed=123)
# Check correspondence: difference in variance can occur due to different poverty
# lines
all.equal(arpr_all$valueByStratum, direct_all_cali$ind[, c("Domain","Head_Count")])
all.equal(arpr_all$valueByStratum$value/100, direct_all_cali$ind$Head_Count)
all.equal(arpr_all_cali$varByStratum, direct_all_cali$MSE[, c("Domain","Head_Count")])
all.equal(arpr_all_cali$varByStratum$var/10000,
direct_all_cali$MSE$Head_Count)
gini_all_cali <- variance("eqIncome",
weights="rb050",
data=eusilc,
breakdown = "db040",
bootType="calibrate",
indicator=gini_all,
R=50,
X = as.matrix(eusilc$age),
seed=123)
# Check correspondence
all.equal(gini_all$valueByStratum, direct_all_cali$ind[, c("Domain","Gini")])
all.equal(gini_all_cali$varByStratum, direct_all_cali$MSE[, c("Domain","Gini")])
qsr_all <- qsr(eusilc$eqIncome,breakdown = eusilc$db040, weights=eusilc$rb050)
qsr_all_cali <- variance("eqIncome",
weights="rb050",
data=eusilc,
breakdown = "db040",
bootType="calibrate",
indicator=qsr_all,
R=50,
X = as.matrix(eusilc$age),
seed=123)
# Check correspondence
all.equal(qsr_all$valueByStratum, direct_all_cali$ind[, c("Domain","Quintile_Share")])
all.equal(qsr_all_cali$varByStratum, direct_all_cali$MSE[, c("Domain","Quintile_Share")])
# 3. without weights but with sampling domains (naive bootstrap)
direct_wow_naive <- direct(y="eqIncome",
smp_data=eusilc,
smp_domains="db040",
weights= NULL ,
# without weights
threshold = 10848.8,
# with weights
#threshold=10859.24,
var=TRUE,
bootType = "naive",
X = NULL,
totals = NULL,
B=50,
seed=123,
na.rm=TRUE)
# laeken indicators that fit to above setting
arpr_wow <- arpr(eusilc$eqIncome, breakdown = eusilc$db040, weights=NULL)
arpr_wow_naive <- variance("eqIncome",
weights=NULL,
data=eusilc,
breakdown = "db040",
indicator=arpr_wow,
R=50,
bootType="naive",
X = NULL,
seed=123)
# Check correspondence: difference in variance can occur due to different poverty
# lines
all.equal(arpr_wow$valueByStratum, direct_wow_naive$ind[, c("Domain","Head_Count")])
all.equal(arpr_wow_naive$varByStratum, direct_wow_naive$MSE[, c("Domain","Head_Count")])
gini_wow <- gini(eusilc$eqIncome, breakdown = eusilc$db040, weights=NULL)
gini_wow_naive <- variance("eqIncome",
weights=NULL,
data=eusilc,
breakdown = "db040",
bootType="naive",
indicator=gini_wow,
R=50,
seed=123)
# Check correspondence
all.equal(gini_wow$valueByStratum, direct_wow_naive$ind[, c("Domain","Gini")])
all.equal(gini_wow_naive$varByStratum, direct_wow_naive$MSE[, c("Domain","Gini")])
qsr_wow <- qsr(eusilc$eqIncome,breakdown = eusilc$db040, weights=NULL)
qsr_wow_naive <- variance("eqIncome",
weights=NULL,
data=eusilc,
breakdown = "db040",
bootType="naive",
X = NULL,
indicator=qsr_wow,
R=50,
seed=123)
# Check correspondence
all.equal(qsr_wow$valueByStratum, direct_wow_naive$ind[, c("Domain","Quintile_Share")])
all.equal(qsr_wow_naive$varByStratum, direct_wow_naive$MSE[, c("Domain","Quintile_Share")])
# 4. without weights but with sampling domains (calibrate bootstrap)
direct_wow_cali <- direct(y="eqIncome",
smp_data=eusilc,
smp_domains="db040",
weights=NULL,
# without weights
threshold = 10848.8,
# with weights
#threshold=10859.24,
var=TRUE,
bootType = "calibrate",
X = as.matrix(eusilc$age),
totals = NULL,
B=50,
seed=123,
na.rm=TRUE)
# laeken indicators that fit to above setting
arpr_wow_cali <- variance("eqIncome",
weights=NULL,
data=eusilc,
breakdown = "db040",
indicator=arpr_wow,
R=50,
bootType="calibrate",
X = as.matrix(eusilc$age),
seed=123)
# Check correspondence: difference in variance can occur due to different poverty
# lines
all.equal(arpr_wow$valueByStratum, direct_wow_cali$ind[, c("Domain","Head_Count")])
all.equal(arpr_wow_cali$varByStratum, direct_wow_cali$MSE[, c("Domain","Head_Count")])
gini_wow_cali <- variance("eqIncome",
weights=NULL,
data=eusilc,
breakdown = "db040",
bootType="calibrate",
indicator=gini_wow,
R=50,
X = as.matrix(eusilc$age),
seed=123)
# Check correspondence
all.equal(gini_wow$valueByStratum, direct_wow_cali$ind[, c("Domain","Gini")])
all.equal(gini_wow_cali$varByStratum, direct_wow_cali$MSE[, c("Domain","Gini")])
qsr_wow_cali <- variance("eqIncome",
weights=NULL,
data=eusilc,
breakdown = "db040",
bootType="calibrate",
indicator=qsr_wow,
R=50,
X = as.matrix(eusilc$age),
seed=123)
# Check correspondence
all.equal(qsr_wow$valueByStratum, direct_wow_cali$ind[, c("Domain","Quintile_Share")])
all.equal(qsr_wow_cali$varByStratum, direct_wow_cali$MSE[, c("Domain","Quintile_Share")])
# Test ebp Quintile Share Ratio
qsr_laeken <- qsr(eusilc$eqIncome,breakdown = NULL, weights=NULL)
qsr_function2(eusilc$eqIncome)
qsr_function3(eusilc$eqIncome)
qsr_function2 <- function(y) {
I1<-(y <= quantile(y,0.2,type = 7))
I2<-(y > quantile(y,0.8,type = 7))
cat("anzk",sum(I1), "anzg",sum(I2))
(sum(1*y[I2])/sum(rep(1, length(y)))) /
(sum(1*y[I1])/sum(rep(1, length(y))))
}
qsr_function3 <- function(x) {
weights <- rep.int(1, length(x))
q <- incQuintile(x, weights)
iq1 <- x <= q[1]
iq4 <- x > q[2]
(sum(weights[iq4] * x[iq4])/sum(weights[iq4]))/(sum(weights[iq1] *
x[iq1])/sum(weights[iq1]))
}
# 5. With the data from the package emdi
data("eusilcA_smp")
direct_emdi <- direct(y="eqIncome",
smp_data=eusilcA_smp,
smp_domains="district",
weights= NULL ,
# without weights
threshold = 10848.8,
# with weights
#threshold=10859.24,
var=TRUE,
bootType = "naive",
X = NULL,
totals = NULL,
B=5,
seed=123,
na.rm=TRUE)
# laeken indicators that fit to above setting
eusilcA_smp$district <- droplevels(eusilcA_smp$district)
arpr_emdi <- arpr(eusilcA_smp$eqIncome, breakdown = eusilcA_smp$district, weights=NULL)
arpr_emdi_naive <- variance("eqIncome",
weights=NULL,
data=eusilcA_smp,
breakdown = "district",
indicator=arpr_emdi,
R=50,
bootType="naive",
X = NULL,
seed=123)
# Check correspondence: difference in variance can occur due to different poverty
# lines
all.equal(arpr_wow$valueByStratum, direct_wow_naive$ind[, c("Domain","Head_Count")])
all.equal(arpr_wow_naive$varByStratum, direct_wow_naive$MSE[, c("Domain","Head_Count")])
gini_emdi <- gini(eusilcA_smp$eqIncome, breakdown = eusilcA_smp$district, weights=NULL)
gini_emdi_naive <- variance("eqIncome",
weights=NULL,
data=eusilcA_smp,
breakdown = "district",
bootType="naive",
indicator=gini_emdi,
R=5,
seed=123)
gini_comp <- data.frame(Domain = direct_emdi$ind$Domain,
Gini_emdi = direct_emdi$ind$Gini,
Gini_laeken = gini_emdi$valueByStratum$value,
Gini_MSE_emdi = direct_emdi$MSE$Gini,
Gini_MSE_laeken = gini_emdi_naive$varByStratum$var)
# Check correspondence
all.equal(gini_wow$valueByStratum, direct_wow_naive$ind[, c("Domain","Gini")])
all.equal(gini_wow_naive$varByStratum, direct_wow_naive$MSE[, c("Domain","Gini")])
qsr_emdi <- qsr(eusilcA_smp$eqIncome,breakdown = eusilcA_smp$district, weights=NULL)
qsr_emdi_naive <- variance("eqIncome",
weights=NULL,
data=eusilcA_smp,
breakdown = "district",
bootType="naive",
X = NULL,
indicator=qsr_emdi,
R=50,
seed=123)
# Check correspondence
all.equal(qsr_wow$valueByStratum, direct_wow_naive$ind[, c("Domain","Quintile_Share")])
all.equal(qsr_wow_naive$varByStratum, direct_wow_naive$MSE[, c("Domain","Quintile_Share")])
SoerenPannier/emdi documentation built on Nov. 2, 2023, 7:54 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
## New Function direct
library(laeken)
library(emdi)
data(eusilc)
# 1. with weights and sampling domains (naive bootstrap)
direct_all_naive <- direct(y = "eqIncome",
smp_data = eusilc,
smp_domains = "db040",
weights = "rb050",
# without weights
# with weights
#threshold=10859.24,
threshold = function(y,weights){0.6 * weightedMedian(x = y, weights = weights)},
var=TRUE,
#bootType = "naive",
boot_type = "naive",
X = NULL,
totals = NULL,
B=50,
seed=123,
na.rm=TRUE)
# laeken indicators that fit to above setting
arpr_all <- arpr(eusilc$eqIncome, breakdown = eusilc$db040, weights=eusilc$rb050)
arpr_all_naive <- variance("eqIncome",
weights="rb050",
data=eusilc,
breakdown = "db040",
indicator=arpr_all,
R=50,
bootType="naive",
X = NULL,
seed=123)
# Check correspondence: difference in variance can occur due to different poverty
# lines
all.equal(arpr_all$valueByStratum, direct_all_naive$ind[, c("Domain","Head_Count")])
all.equal(arpr_all_naive$varByStratum, direct_all_naive$MSE[, c("Domain","Head_Count")])
all.equal(arpr_all_naive$varByStratum$var/10000, direct_all_naive$MSE[, c("Domain","Head_Count")][,2])
gini_all <- gini(eusilc$eqIncome, breakdown = eusilc$db040, weights=eusilc$rb050)
gini_all_naive <- variance("eqIncome",
weights="rb050",
data=eusilc,
breakdown = "db040",
bootType="naive",
indicator=gini_all,
R=50,
seed=123)
# Check correspondence
all.equal(gini_all$valueByStratum, direct_all_naive$ind[, c("Domain","Gini")])
all.equal(gini_all_naive$varByStratum, direct_all_naive$MSE[, c("Domain","Gini")])
qsr_all <- qsr(eusilc$eqIncome,breakdown = eusilc$db040, weights=eusilc$rb050)
qsr_all_naive <- variance("eqIncome",
weights="rb050",
data=eusilc,
breakdown = "db040",
bootType="naive",
X = NULL,
indicator=qsr_all,
R=50,
seed=123)
# Check correspondence
all.equal(qsr_all$valueByStratum, direct_all_naive$ind[, c("Domain","Quintile_Share")])
all.equal(qsr_all_naive$varByStratum, direct_all_naive$MSE[, c("Domain","Quintile_Share")])
# 2. with weights and sampling domains (calibrate bootstrap)
direct_all_cali <- direct(y="eqIncome",
smp_data=eusilc,
smp_domains="db040",
weights="rb050",
# without weights
#threshold = 10848.8,
# with weights
threshold=10859.24,
var=TRUE,
bootType = "calibrate",
X = as.matrix(eusilc$age),
totals = NULL,
B=50,
seed=123,
na.rm=TRUE)
direct_all_cali <- direct(y="eqIncome",
smp_data=eusilc,
smp_domains="db040",
weights="rb050",
# without weights
#threshold = 10848.8,
# with weights
threshold=10859.24,
var=TRUE,
boot_type = "calibrate",
X_calib = as.matrix(eusilc$age),
totals = NULL,
B=50,
seed=123,
na.rm=TRUE)
# laeken indicators that fit to above setting
arpr_all_cali <- variance("eqIncome",
weights="rb050",
data=eusilc,
breakdown = "db040",
indicator=arpr_all,
R=50,
bootType="calibrate",
X = as.matrix(eusilc$age),
seed=123)
# Check correspondence: difference in variance can occur due to different poverty
# lines
all.equal(arpr_all$valueByStratum, direct_all_cali$ind[, c("Domain","Head_Count")])
all.equal(arpr_all$valueByStratum$value/100, direct_all_cali$ind$Head_Count)
all.equal(arpr_all_cali$varByStratum, direct_all_cali$MSE[, c("Domain","Head_Count")])
all.equal(arpr_all_cali$varByStratum$var/10000,
direct_all_cali$MSE$Head_Count)
gini_all_cali <- variance("eqIncome",
weights="rb050",
data=eusilc,
breakdown = "db040",
bootType="calibrate",
indicator=gini_all,
R=50,
X = as.matrix(eusilc$age),
seed=123)
# Check correspondence
all.equal(gini_all$valueByStratum, direct_all_cali$ind[, c("Domain","Gini")])
all.equal(gini_all_cali$varByStratum, direct_all_cali$MSE[, c("Domain","Gini")])
qsr_all <- qsr(eusilc$eqIncome,breakdown = eusilc$db040, weights=eusilc$rb050)
qsr_all_cali <- variance("eqIncome",
weights="rb050",
data=eusilc,
breakdown = "db040",
bootType="calibrate",
indicator=qsr_all,
R=50,
X = as.matrix(eusilc$age),
seed=123)
# Check correspondence
all.equal(qsr_all$valueByStratum, direct_all_cali$ind[, c("Domain","Quintile_Share")])
all.equal(qsr_all_cali$varByStratum, direct_all_cali$MSE[, c("Domain","Quintile_Share")])
# 3. without weights but with sampling domains (naive bootstrap)
direct_wow_naive <- direct(y="eqIncome",
smp_data=eusilc,
smp_domains="db040",
weights= NULL ,
# without weights
threshold = 10848.8,
# with weights
#threshold=10859.24,
var=TRUE,
bootType = "naive",
X = NULL,
totals = NULL,
B=50,
seed=123,
na.rm=TRUE)
# laeken indicators that fit to above setting
arpr_wow <- arpr(eusilc$eqIncome, breakdown = eusilc$db040, weights=NULL)
arpr_wow_naive <- variance("eqIncome",
weights=NULL,
data=eusilc,
breakdown = "db040",
indicator=arpr_wow,
R=50,
bootType="naive",
X = NULL,
seed=123)
# Check correspondence: difference in variance can occur due to different poverty
# lines
all.equal(arpr_wow$valueByStratum, direct_wow_naive$ind[, c("Domain","Head_Count")])
all.equal(arpr_wow_naive$varByStratum, direct_wow_naive$MSE[, c("Domain","Head_Count")])
gini_wow <- gini(eusilc$eqIncome, breakdown = eusilc$db040, weights=NULL)
gini_wow_naive <- variance("eqIncome",
weights=NULL,
data=eusilc,
breakdown = "db040",
bootType="naive",
indicator=gini_wow,
R=50,
seed=123)
# Check correspondence
all.equal(gini_wow$valueByStratum, direct_wow_naive$ind[, c("Domain","Gini")])
all.equal(gini_wow_naive$varByStratum, direct_wow_naive$MSE[, c("Domain","Gini")])
qsr_wow <- qsr(eusilc$eqIncome,breakdown = eusilc$db040, weights=NULL)
qsr_wow_naive <- variance("eqIncome",
weights=NULL,
data=eusilc,
breakdown = "db040",
bootType="naive",
X = NULL,
indicator=qsr_wow,
R=50,
seed=123)
# Check correspondence
all.equal(qsr_wow$valueByStratum, direct_wow_naive$ind[, c("Domain","Quintile_Share")])
all.equal(qsr_wow_naive$varByStratum, direct_wow_naive$MSE[, c("Domain","Quintile_Share")])
# 4. without weights but with sampling domains (calibrate bootstrap)
direct_wow_cali <- direct(y="eqIncome",
smp_data=eusilc,
smp_domains="db040",
weights=NULL,
# without weights
threshold = 10848.8,
# with weights
#threshold=10859.24,
var=TRUE,
bootType = "calibrate",
X = as.matrix(eusilc$age),
totals = NULL,
B=50,
seed=123,
na.rm=TRUE)
# laeken indicators that fit to above setting
arpr_wow_cali <- variance("eqIncome",
weights=NULL,
data=eusilc,
breakdown = "db040",
indicator=arpr_wow,
R=50,
bootType="calibrate",
X = as.matrix(eusilc$age),
seed=123)
# Check correspondence: difference in variance can occur due to different poverty
# lines
all.equal(arpr_wow$valueByStratum, direct_wow_cali$ind[, c("Domain","Head_Count")])
all.equal(arpr_wow_cali$varByStratum, direct_wow_cali$MSE[, c("Domain","Head_Count")])
gini_wow_cali <- variance("eqIncome",
weights=NULL,
data=eusilc,
breakdown = "db040",
bootType="calibrate",
indicator=gini_wow,
R=50,
X = as.matrix(eusilc$age),
seed=123)
# Check correspondence
all.equal(gini_wow$valueByStratum, direct_wow_cali$ind[, c("Domain","Gini")])
all.equal(gini_wow_cali$varByStratum, direct_wow_cali$MSE[, c("Domain","Gini")])
qsr_wow_cali <- variance("eqIncome",
weights=NULL,
data=eusilc,
breakdown = "db040",
bootType="calibrate",
indicator=qsr_wow,
R=50,
X = as.matrix(eusilc$age),
seed=123)
# Check correspondence
all.equal(qsr_wow$valueByStratum, direct_wow_cali$ind[, c("Domain","Quintile_Share")])
all.equal(qsr_wow_cali$varByStratum, direct_wow_cali$MSE[, c("Domain","Quintile_Share")])
# Test ebp Quintile Share Ratio
qsr_laeken <- qsr(eusilc$eqIncome,breakdown = NULL, weights=NULL)
qsr_function2(eusilc$eqIncome)
qsr_function3(eusilc$eqIncome)
qsr_function2 <- function(y) {
I1<-(y <= quantile(y,0.2,type = 7))
I2<-(y > quantile(y,0.8,type = 7))
cat("anzk",sum(I1), "anzg",sum(I2))
(sum(1*y[I2])/sum(rep(1, length(y)))) /
(sum(1*y[I1])/sum(rep(1, length(y))))
}
qsr_function3 <- function(x) {
weights <- rep.int(1, length(x))
q <- incQuintile(x, weights)
iq1 <- x <= q[1]
iq4 <- x > q[2]
(sum(weights[iq4] * x[iq4])/sum(weights[iq4]))/(sum(weights[iq1] *
x[iq1])/sum(weights[iq1]))
}
# 5. With the data from the package emdi
data("eusilcA_smp")
direct_emdi <- direct(y="eqIncome",
smp_data=eusilcA_smp,
smp_domains="district",
weights= NULL ,
# without weights
threshold = 10848.8,
# with weights
#threshold=10859.24,
var=TRUE,
bootType = "naive",
X = NULL,
totals = NULL,
B=5,
seed=123,
na.rm=TRUE)
# laeken indicators that fit to above setting
eusilcA_smp$district <- droplevels(eusilcA_smp$district)
arpr_emdi <- arpr(eusilcA_smp$eqIncome, breakdown = eusilcA_smp$district, weights=NULL)
arpr_emdi_naive <- variance("eqIncome",
weights=NULL,
data=eusilcA_smp,
breakdown = "district",
indicator=arpr_emdi,
R=50,
bootType="naive",
X = NULL,
seed=123)
# Check correspondence: difference in variance can occur due to different poverty
# lines
all.equal(arpr_wow$valueByStratum, direct_wow_naive$ind[, c("Domain","Head_Count")])
all.equal(arpr_wow_naive$varByStratum, direct_wow_naive$MSE[, c("Domain","Head_Count")])
gini_emdi <- gini(eusilcA_smp$eqIncome, breakdown = eusilcA_smp$district, weights=NULL)
gini_emdi_naive <- variance("eqIncome",
weights=NULL,
data=eusilcA_smp,
breakdown = "district",
bootType="naive",
indicator=gini_emdi,
R=5,
seed=123)
gini_comp <- data.frame(Domain = direct_emdi$ind$Domain,
Gini_emdi = direct_emdi$ind$Gini,
Gini_laeken = gini_emdi$valueByStratum$value,
Gini_MSE_emdi = direct_emdi$MSE$Gini,
Gini_MSE_laeken = gini_emdi_naive$varByStratum$var)
# Check correspondence
all.equal(gini_wow$valueByStratum, direct_wow_naive$ind[, c("Domain","Gini")])
all.equal(gini_wow_naive$varByStratum, direct_wow_naive$MSE[, c("Domain","Gini")])
qsr_emdi <- qsr(eusilcA_smp$eqIncome,breakdown = eusilcA_smp$district, weights=NULL)
qsr_emdi_naive <- variance("eqIncome",
weights=NULL,
data=eusilcA_smp,
breakdown = "district",
bootType="naive",
X = NULL,
indicator=qsr_emdi,
R=50,
seed=123)
# Check correspondence
all.equal(qsr_wow$valueByStratum, direct_wow_naive$ind[, c("Domain","Quintile_Share")])
all.equal(qsr_wow_naive$varByStratum, direct_wow_naive$MSE[, c("Domain","Quintile_Share")])
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.