# Solutions for Exercise 3: Introduction to 
# Descriptive and Inferential Statistics with R


# 1
# cardat.csv
dat<-read.csv(file.choose())
head(dat)
names(dat)

# 2
mean(dat$purc)
tapply(dat$twoyr, dat$type, mean, tr=.2)
tapply(dat$oneyr, list(dat$type,dat$sex), mean)

tapply(dat$twoyr, dat$type, describe, trim=.2)

describeBy(dat$twoyr,dat$type,trim=.2)
describeBy(dat$oneyr, list(dat$type,dat$sex),trim=.2)

hist(dat$purc)


# 3
library(psych)
describe(dat)

# 4
tab<-table(dat$type)
barplot(tab)

# 5
leveneTest(dat$twoyr,dat$sex)
dat2<-subset(dat,type=="car" | type=="truck")
library(car)
leveneTest(dat$twoyr[dat$type=="car"|dat$type=="truck"],dat$type[dat$type=="car"|dat$type=="truck"])
leveneTest(dat2$twoyr,dat2$type)
hist(dat$twoyr[dat$sex=="male"])
hist(dat$twoyr[dat$sex=="female"])
tapply(dat$twoyr,dat$sex,shapiro.test)
hist(dat2$twoyr[dat2$type=="car"])
hist(dat2$twoyr[dat2$type=="truck"])
tapply(dat$twoyr,dat$type,shapiro.test)

t.test(twoyr~sex, data=dat)

t.test(twoyr~type, data=dat2)
#or
t.test(dat$twoyr[dat$type=="car"],dat$twoyr[dat$type=="truck"])

library(WRS2)
yuen(twoyr~sex,data=dat)
yuen(twoyr~type,data=dat2)

# 6
t.test(dat$purc,dat$oneyr, paired=TRUE)

# 7
mod1<-lm(twoyr~purc,data=dat)
summary(mod1)
cutoff <- 4/(length(dat$twoyr)-1)
plot(mod1, which=4, cook.levels=cutoff)
dat3<-dat[-26,]
mod2<-lm(twoyr~purc,data=dat3)
summary(mod2)

# 8
mod3<-lm(twoyr ~ sex + purc, data=dat)
summary(mod3)