#### a silly non-parametric approach that illustrates what we're doing
SLID <- read.table('http://math.uttyler.edu/nathan/classes/statistics/data/slid.data',header=TRUE)
SLID[1:10,]
attach(SLID)
plot(education,wages)


boxplot(wages[education==7],wages[education==8],wages[education==9],
        wages[education==10],wages[education==11],wages[education==12],
        wages[education==13],wages[education==14],wages[education==15],
        wages[education==16],wages[education==17],wages[education==18],
        wages[education==19],wages[education==20],
        main='Wages at various educaiton levels',
        xlab='Education',
        names=c('7','8','9','10','11','12','13','14','15','16','17','18','19','20'))

### note there is an entire (wages) population for each level of
### education -- this is important to grasp
      
plot(education,wages)
for (i in 0:20) {
  points(i,mean(na.omit(wages[education==i])),pch=19,col='red')}
lines(0:20,sapply(0:20,function (i) {mean(na.omit(wages[education==i]))}),col='red')

detach(SLID)


### we'll beat this data set to death in this class, you'll eventually
### grow to hate it, so consider beating it to death a small pleasure!
### from this point on we're parametric again
Prestige <- read.table('http://math.uttyler.edu/nathan/classes/statistics/data/prestige.data',header=TRUE)
Prestige[1:10,]
attach(Prestige)

plot(education,income)
names(Prestige)

mod.1 <- lm(income ~ education)
abline(mod.1,col='red')
summary(mod.1)
### what do all of these numbers mean?

abline(h=mean(income[income<15000]),col='green')
### is red better than green?