###########################################################################
# PUBLG100: Introduction to Quantitative Methods
#
# Week 6 Solutions: Assumptions and Violations of Assumptions
#
#

## ------------------------------------------------------------------------
rm(list = ls())

## ----message = FALSE-----------------------------------------------------
library(lmtest)
library(sandwich)
library(texreg)
library(dplyr)

## ------------------------------------------------------------------------
MA_schools <- read.csv("MA_Schools.csv")

## ------------------------------------------------------------------------
model1 <- lm(score8 ~ english, data = MA_schools)
summary(model1)

## ------------------------------------------------------------------------
plot(score8 ~ english, data = MA_schools)
abline(model1, col = "red")

## ------------------------------------------------------------------------
summary(select(MA_schools, score8, stratio, english, lunch, income))

## ------------------------------------------------------------------------
cor(select(MA_schools, score8, stratio, english, lunch, income))

## ------------------------------------------------------------------------
cor(select(MA_schools, score8, stratio, english, lunch, income), 
    use = "pairwise.complete.obs")

## ------------------------------------------------------------------------
model2 <- lm(score8 ~ english + lunch + income, data = MA_schools)
summary(model2)

## ------------------------------------------------------------------------
screenreg(list(model1, model2))

## ------------------------------------------------------------------------
plot(fitted(model2), residuals(model2))

## ------------------------------------------------------------------------
bptest(model2)