sae

#sae

wilayah<-podes[,c(1:9)]
head(wilayah)
str_to_lower(wilayah$R103N)
str_replace(str_to_lower(wilayah$R103N),' ','')
wilayah$kec<-str_replace(str_to_lower(wilayah$R103N),' ','')
sek_desa<-sek_desa
sekdesa<-sek_desa
sekdesa$kec <- gsub(' ','',str_replace(str_to_lower(sekdesa$Kecamatan),'kec. ',''))
aggregate(R104 ~ R103N, wilayah[ !duplicated(wilayah$R103N),], FUN = function(x){ NROW(x)})


ggscatter(test.2, x = "xCap", y = "Minimarket.swalayan", 
          add = "reg.line", conf.int = TRUE, 
          cor.coef = TRUE, cor.method = "pearson",
          xlab = "Toko", ylab = "rata exp_cap")

####################### Aggregate menggunakan fungsi ###

 data("mtcars")
 library(plyr)
 ddply(mtcars,"cyl",function(x) cor(x$hp,x$wt))
################################################
###############################################

tDesa <- read.csv('jumlah sarana ekonomi desa.xlsx - Sheet1.csv', header=T)
kab<-aggregate(.~Id.Prov+Provinsi+Id.Kab+Kabupaten, tDesa[,c(3:6,11:21)], sum)
kab <- kab[ order(kab$Id.Kab),]

expcap<-read.dbf('../Documents/datalab/SUSENAS/SUSENAS Maret 2018/Data kor 2018/kor18ind_revisi_diseminasi.dbf')[,c(1,2,171,172)]
expcap$R102 <- paste0(expcap$R101,sprintf('%02d',expcap$R102))
expcap$mul <- expcap$FWT * expcap$EXP_CAP


exKab<-merge(
aggregate(mul ~ R101+R102, expcap, sum),
aggregate(FWT ~ R101+R102, expcap, sum),
by=c('R101','R102'))

# merge variance sample expcap dengan exKab
exKab<-merge(
exKab,
ddply(expcap,c('R101','R102'),function(x) var(x$EXP_CAP)),
by=c('R101','R102'))


exKab$xCap <- exKab$mul/exKab$FWT


exPro<-merge(
aggregate(mul ~ R101, expcap, sum),
aggregate(FWT ~ R101, expcap, sum),
by=c('R101'))

# uji dengan pengguna listrik rumahtangga

Dmukim <- read.csv('Indonesia_podes_2018_pemukiman.csv')
Dmukim$R102 <- paste0(Dmukim$R101,sprintf('%02d',Dmukim$R102))
Kmukim <- aggregate( .~ R101+R101N+R102+R102N, Dmukim[,c(1:4,9:11)],sum)
Kmukim <- Kmukim[ order(Kmukim$R102),]
Kmukim$Tot <- rowSums(Kmukim[,c(5:7)])
Kmukim$totA12 <- rowSums(Kmukim[,c(5,6)])
Kmukim$percent <- 100*Kmukim$totA12/Kmukim$Tot

test.3<-merge(exKab[ exKab$R101 %in% c(11:36),], Kmukim, by.x=c('R101','R102'))

df <- data.frame()
for(i in unique(test.3$R101)){
	res<-cor.test(test.3[test.3$R101 == i,]$xCap,test.3[test.3$R101 == i,]$percent)
	test <- ifelse(res$p.value < 0.05, paste0('ok'),paste0('not-ok'))
	df<-rbind(df,data.frame(R101=i, koef.cor = as.vector(res$estimate), p.value=res$p.value, test=test))
}

test.2<-merge(exKab[ exKab$R101 %in% c(11:36),], kab, by.x=c('R101','R102'), by.y=c('Id.Prov','Id.Kab'))

c("Kelompok.pertokoan","Minimarket.swalayan","Pasar.dengan.bangunan.permanen","Pasar.dengan.bangunan.semi.permanen","Pasar.tanpa.bangunan","Penginapan","Restoran.rumah.makan","Toko.warung.kelontong","Toko.warung.kelontong.yang.menjual.bahan.pangan","Warung.kedai.makanan.minuman")

df <- data.frame()
for( j in colnames(test.2[8:18]) ){
	for( i in c("11","12","13","14","15","16","17","18","19","21","31","32","33","34","35","36")){
		l <- cor(test.2[ test.2$R101 == i,]$xCap,test.2[ test.2$R101 == i,j])
		df<-rbind(df,data.frame(R101=i, cor.coef=l, param=j))
	}
}

df.1 <- data.frame()
for( j in colnames(test.2[8:18]) ){
	for( i in unique(test.2$R101)){
		l <- cor(test.2[ test.2$R101 == i,]$xCap,test.2[ test.2$R101 == i,j])
		df.1<-rbind(df,data.frame(R101=i, cor.coef=l, param=j))
	}
}


lm(xCap ~ 0+Hotel, data=test.2[test.2$R101,])

#log

for( i in c("11","12","13","14","15","16","17","18","19","21","31","32","33","34","35","36")){
	l <- cor(log(test.2[ test.2$R101 == i,]$xCap),test.2[test.2$R101 == i,]$Minimarket.swalayan)
	print(l)
}