# Ucitavanje podataka
creditdata = read.csv('creditdata.csv')
summary(creditdata)
# Priprema podataka
creditdata$education = factor(creditdata$education,levels = c(1,2,3),labels = c('elementary','secondary', 'university'))
creditdata$marriage = factor(creditdata$marriage,levels = c(1,2),labels = c('single','married'))
creditdata$apartment = factor(creditdata$apartment,levels = c(1,2),labels = c('rent','own'))
creditdata$default = factor(creditdata$default,levels = c(0,1),labels = c(FALSE,TRUE))
summary(creditdata)
require(nortest)
lillie.test(creditdata$income)
lillie.test(creditdata$income[creditdata$education=='elementary'])
lillie.test(creditdata$income[creditdata$education=='secondary'])
lillie.test(creditdata$income[creditdata$education=='university'])
hist(creditdata$income[creditdata$education=='elementary'])
hist(creditdata$income[creditdata$education=='secondary'])
hist(creditdata$income[creditdata$education=='university'])
# Testiranje homogenosti varijance uzoraka Bartlettovim testom
bartlett.test(creditdata$income ~ creditdata$education)
var((creditdata$income[creditdata$education=='elementary']))
var((creditdata$income[creditdata$education=='secondary']))
var((creditdata$income[creditdata$education=='university']))
# Graficki prikaz podataka
boxplot(creditdata$income ~ creditdata$education)
# Test
a = aov(creditdata$income ~ creditdata$education)
summary(a)
# Linearni model
model = lm(income ~ education, data = creditdata)
summary(model)
anova(model)
# if you are trying to convert the data from R to python, https://www.mit.edu/~amidi/teaching/data-science-tools/conversion-guide/r-python-data-manipulation/ is a good resource
# if you are looking for translating the code, https://towardsdatascience.com/essential-guide-to-translating-between-python-and-r-7cb18b786e5d seems to have comprehsensive information