library(modeest)
library(mice)
getwd()
setwd("/home/vantu/Thesis")


data <- read.csv("insurance.csv")
original <- data
head(data)

# num of row and column

print(c("num of row ",nrow(data)))

print(c("num of row ",ncol(data)))

# count male and female

data$sex<- factor(data$sex,labels = c("female","male"))
table(data$sex)

# mode of the target attribute

ux<-  unique(data$sex)
ux<-ux[which.max(tabulate(match(data$sex,ux)))]

ux

# % of missing data

data[sample(1:nrow(data), 1000), "sex"] <- NA   # insert na
sapply(data, function(x) sum(is.na(x)))
na_index=which(is.na(data$sex),arr.ind = TRUE)  # pop_out na index
miss_data<- (length(na_index)/nrow(data))*100;  # count % of missing value
print(miss_data)


# normal distribution ...of age

# dnorm

x<-original$age

y<- dnorm(x,mean = mean(x),sd = sd(x))
plot(x,y,main = "Normal distribution of age",xlab = "x",ylab = "probability Density",las=1)
##abline() use to draw a vertical line through mean of x
abline(v=mean(x))

##by pnorm()

y<- pnorm(x,mean = mean(x),sd = sd(x))
plot(x,y)

##by rnorm()
y<- rnorm(x,mean = mean(x),sd = sd(x))

plot(x,y)

##qnorm()
y<- qnorm(x,mean = mean(x),sd = sd(x))

plot(x,y)