# Rafael Almeida — ISM4542 Statistical Programming for Business Analytics
# Module 6: multiple regression on Cadillac listings
library(rio)
cars <- import("4542 Module 6 Assignment Data.xlsx")
cadillac <- subset(cars, tolower(make) == "cadillac")
cadillac$cylinders <- factor(cadillac$cylinders)
cadillac$condition <- factor(cadillac$condition)
cadillac$age <- as.numeric(cadillac$age)
cadillac$age2 <- cadillac$age^2
str(cadillac)
table(cadillac$make)
price_model <- lm(price ~ odometer + age + age2 + condition + cylinders, data = cadillac)
summary(price_model)
confint(price_model)
par(mfrow = c(2, 2)); plot(price_model); par(mfrow = c(1, 1))
new_car <- data.frame(odometer = 77242, age = 7, age2 = 7^2,
  condition = factor("excellent", levels = levels(cadillac$condition)),
  cylinders = factor(8, levels = levels(cadillac$cylinders)))
predict(price_model, newdata = new_car, interval = "prediction")