##зареждане на нужните данни##



housing <-read.csv("http://www.observatoria.eu/boyan/sites/default/files/bliss_m3.csv", sep=",")
dict_housing <-read.csv("http://www.observatoria.eu/boyan/sites/default/files/dict_BLISS_m3.csv", sep=",")
blissm1 <- read.csv ("http://www.observatoria.eu/boyan/sites/default/files/bliss_m1_1.csv", sep=",")
dict_blissm1 <-read.csv ("http://www.observatoria.eu/boyan/sites/default/files/dict_BLISS_m1.csv", sep=",")
blissm10<-read.csv("http://www.observatoria.eu/boyan/sites/default/files/bliss_m10.csv", sep=",")
dict_blissm10<-read.csv("http://www.observatoria.eu/boyan/sites/default/files/dict_BLISS_m10.csv", sep=",")
blissm4<-read.csv("http://www.observatoria.eu/boyan/sites/default/files/bliss_m4.csv",sep=",")
dict_blissm4<-read.csv("http://www.observatoria.eu/boyan/sites/default/files/dict_BLISS_m4.csv", sep=",")
wt_cs_ind<-read.csv("http://www.observatoria.eu/boyan/sites/default/files/wt_BLISS_ms_cs_ind.csv", sep=",")
wt_cs_hh<-read.csv("http://www.observatoria.eu/boyan/sites/default/files/wt_BLISS_ms_cs_hh.csv", sep=",")

## тези два модула не са необходими##

hh <-read.csv("C:/Users/bzahariev/Documents/Training/open data/bliss/bliss_hh.csv", sep=";")
hhm <-read.csv("C:/Users/bzahariev/Documents/Training/open data/bliss/bliss_hhm.csv", sep=";")


##модул m3 е осветен на жилището, module m3 is about the dwelling##


##m3_q3 e броят на стаите, с които домакинството разполага, остава да пресметнем колко стаи трябва да имат според дефиницията на Евростат. m3_q3 is the number of rooms. we have to calculate the number of rooms needed##

##blissm1$m1_q2 = relation to the head, 1=head, 2=spouse, blissm1$m1_q4a_age##
##blissm1$m1_q3 = sex, blissm1$m1_q2 = relation to the head, blissm1$m1_q4a_age, blissm1$m1_q5 = marital status, 1=single, 2=married, 3=living together##

blissm1 <- rename(blissm1, c(HH_ID="household"))
names(blissm1)[names(blissm1)=="HH_ID"] <- "household"
names(blissm1)[names(blissm1)=="HHM_ID"] <- "person"

##по-горе е даден пример за преименуване на променливи##

marital<-subset(blissm1,select=c(household, person,m1_q5))

##Тези команди не са необходими повече##
hhm<-merge(hhm,marital, by=c("household","person"), all.hhm=T, sort =T)
hhm2<-merge(hhm,subset(blissm1,select=c(household, person,m1_q5)), by=c("household","person"), all.hhm=T, sort =T)


#one room for the household;
#one room per couple in the household;
#one room for each single person aged 18 or more;
#one room per pair of single people of the same gender between 12 and 17 years of age;
#one room for each single person between 12 and 17 years of age and not included in the previous category;
#one room per pair of children under 12 years of age.

##създаване на променлива за броя на стаите, creating a variable that will hold the data of rooms needed##

blissm1$rooms<-1
housing$rooms<-1

##проверка дали променливата е количествена, checking if age is numeric##
is.numeric(blissm1$m1_q4a_age)
mean(blissm1$m1_q4a_age, na.rm=T)

##Копиране на същия масив с данни с ново име. Copying the data with a new name.##

hhm<-blissm1

## ако някой няма партньор, if somebody is not a couple. Това са само фрагменти от логическата структура на кода, още не са истински команди##
#(hhm$m1_q5 != 2 | hhm$m1_q5 != 3) + 1
#(hhm$m1_q5 = 2 | hhm$m1_q5 = 3) + 0.5
##ако някой попада в определена възрастова група##
#(hhm$m1_q4a_age>11 $ hhm$m1_q4a<18) + 1
#(hhm$m1_q4a_age<12) + ceiling(0.5)

##при децата създаваме категория за деца до 12 години, като за определяне на броя стаи делим на две и закръгляме нагоре##
hhm$children11<-ifelse(hhm$m1_q4a_age<12,1,0)
sum(hhm$children11,na.rm=T)

##при възрастните създаваме категория singe и категория couple, при couple делим на две, при single сумираме, при living together се предполага, че са в едно домакинство, при married може да има проблем##
hhm$single<-ifelse(hhm$m1_q5 != 2 & hhm$m1_q5 !=3 ,1,0)
hhm$couple<-ifelse(hhm$m1_q5 == 2 | hhm$m1_q5 ==3 ,1,0)

##при категорията 12-17 създаваме две променливи - мъже и жени, в рамките на вкяка променлива сумираме и закръгляме нагоре##
hhm$children17m<-ifelse(hhm$m1_q4a_age > 11 & hhm$m1_q4a_age < 18 & hhm$m1_q3 == 1 ,1,0)
hhm$children17f<-ifelse(hhm$m1_q4a_age > 11 & hhm$m1_q4a_age < 18 & hhm$m1_q3 == 2 ,1,0)


hhm1<-subset(hhm, select=c(household, children11:children17f))

##Понякога може да се наложи следвата операция. В случая не се налага.##

#hhm1$children11<-as.numeric(unlist(hhm1$children11))##
#hhm1$children17m<-as.numeric(unlist(hhm1$children17m))##
#hhm1$children17f<-as.numeric(unlist(hhm1$children17f))##
#hhm1$single<-as.numeric(unlist(hhm1$single))##
#hhm1$couple<-as.numeric(unlist(hhm1$couple))##

hhm2<-aggregate(.~household, data=hhm1, sum)

names(housing)[names(housing)=="HH_ID"] <- "household"
names(housing)[names(housing)=="HHM_ID"] <- "person"
intersect(names(housing), names(hhm2))



hhm2$roomsneeded<-1+ceiling(hhm2$children11/2) + hhm2$single + ceiling(hhm2$couple/2) + ceiling(hhm2$children17m/2) + ceiling(hhm2$children17f/2)


hhm2<-merge(hhm2,subset(housing,select=c(household, m3_q3, m3_q5_1, m3_q5_3, m3_q5_6, m3_q5_7)), by="household", all.hhm2=T, sort =T)

names(hhm2)[names(hhm2)=="m3_q3"] <- "roomsavailable"
names(hhm2)[names(hhm2)=="m3_q5_3"] <- "electricity"
names(hhm2)[names(hhm2)=="m3_q5_1"] <- "water"
names(hhm2)[names(hhm2)=="m3_q5_6"] <- "toilet"
names(hhm2)[names(hhm2)=="m3_q5_7"] <- "bathroom"

hhm2$overcrowd<-ifelse((hhm2$roomsavailable - hhm2$roomsneeded)<0,1,0)

blissm1<-merge(blissm1,subset(hhm2,select=c(household, overcrowd, water, electricity, toilet, bathroom)), by="household", all.blissm1=T, sort =T)

##Преименуваме променливите, свързани със здравето##
names(blissm1)[names(blissm1)=="m1_q7"] <- "insurance"
names(blissm1)[names(blissm1)=="m1_q9a"] <- "health"
names(blissm1)[names(blissm1)=="m1_q9c"] <- "disability"

##Изчисляваме променливите, свързани с образованието##
blissm1$primary<-ifelse(blissm1$m1_q14a < 4 & blissm1$m1_q4a_age >18,1,0)
blissm1$dropout<-ifelse(blissm1$m1_q11a==7 & blissm1$m1_q4a_age >7 & blissm1$m1_q4a_age < 16,1,0)

names(blissm10)[names(blissm10)=="HH_ID"] <- "household"
names(blissm10)[names(blissm10)=="m10_q9а"] <- "holiday"
names(blissm10)[names(blissm10)=="m10_q10"] <- "eat"
names(blissm10)[names(blissm10)=="m10_q19"] <- "income"
names(blissm10)[names(blissm10)=="poverty_rate"] <- "mon_poverty"
names(blissm10)[names(blissm10)=="m10_q20"] <- "language"

blissm1<-merge(blissm1,subset(blissm10,select=c(household, holiday, eat, income, income_modif, mon_poverty, language)),by="household",all.blissm1=T, sort=T)

names(blissm4)[names(blissm4)=="HH_ID"] <- "household"
names(blissm4)[names(blissm4)=="m4_q1"] <- "television"
names(blissm4)[names(blissm4)=="m4_q10"] <- "washing"
names(blissm4)[names(blissm4)=="m4_q16"] <- "car"
names(blissm4)[names(blissm4)=="m4_q13"] <- "telephone"
names(blissm4)[names(blissm4)=="m4_q14"] <- "mobile"

install.packages("car", dependencies=T)
install.packages("survey", dependencies=T) 
install.packages("MatrixModels")

require(car)

blissm4$television<-recode(blissm4$television,"0=1; 1:1000=0")
blissm4$washing<-recode(blissm4$washing,"0=1; 1:1000=0")
blissm4$car<-recode(blissm4$car,"0=1; 1:1000=0")
blissm4$telephone<-recode(blissm4$telephone,"0=1; 1:1000=0")
blissm4$mobile<-recode(blissm4$mobile,"0=1; 1:1000=0")


blissm4$deprived<-ifelse((rowSums(subset(blissm4, select=c(television, washing, car)), na.rm=T)<2 & rowSums(subset(blissm4, select=c(telephone, mobile)), na.rm=T)>0) |rowSums(subset(blissm4, select=c(television, washing, car)), na.rm=T)>1, 1,0)
blissm1<-merge(blissm1,subset(blissm4,select=c(household,deprived)),by="household",all.blissm1=T, sort=T)


##трябва да се прекодират insurance(2), disability (1), health (4 or 5) electricity (2), water (2), toilet (2), bathroom (2), holiday(2), eat(2) ##

blissm1$insurance<-recode(blissm1$insurance,"2=1; 1=0")
blissm1$disability<-recode(blissm1$disability,"2=0")
blissm1$electricity<-recode(blissm1$electricity,"2=1; 1=0")
blissm1$water<-recode(blissm1$water,"2=1; 1=0")
blissm1$toilet<-recode(blissm1$toilet,"2=1; 1=0")
blissm1$bathroom<-recode(blissm1$bathroom,"2=1; 1=0")
blissm1$holiday<-recode(blissm1$holiday,"2=1; 1=0")
blissm1$eat<-recode(blissm1$eat,"2=1; 1=0")
blissm1$poorhealth<-ifelse(blissm1$health == 4 | blissm1$health == 5,1,0)
blissm1$toil_bath<-ifelse(blissm1$toilet==1|blissm1$bathroom==1, 1,0)


blissm1$mdpov<-rowSums(subset(blissm1, select = c(insurance, disability, poorhealth, primary, dropout, holiday, eat, deprived, toil_bath, electricity, water, overcrowd)), na.rm=T)
blissm1$mdpoor<-ifelse(blissm1$mdpov>3,1,0)
mean(blissm1$mdpoor,na.rm=T)

names(wt_cs_ind)[names(wt_cs_ind)=="hh_id"] <- "household"
names(wt_cs_ind)[names(wt_cs_ind)=="hhm_id"] <- "person"
names(wt_cs_ind)[names(wt_cs_ind)=="wt_BLISS_ms_cs_ind_final"] <- "weight"
blissm1<-merge(blissm1, subset(wt_cs_ind, select=c(household, person, weight)), by=c("household", "person"), all.blissm1=T, sort=T)

##претеглени хистограми и боксплотове##

require(survey)
design <- svydesign(id = ~1, strata = NULL, weights = ~weight, data = blissm1)
svyboxplot(income~mdpoor,design)
svyhist(~income, design)