# generate descriptive stats and graphs on weight, height

# get basic histogram of weight
hist(vecheight, main="UM Football WEIGHT Distribution",
       xlab="pounds", ylab="frequency")

# generate relevant summary statistics
summary(vecweight)
cat("std dev=",sd(vecweight),"\n")

# get weight as relative frequency on histogram
wt <- vecweight/length(vecweight) 
hist(wt, main="UM Football WEIGHT Distribution",
       xlab="pounds", ylab="rel frequency")
lines(density(wt))   # this gets a profile line on histogram

# generate another displays of WEIGHT
cat("TEAM WEIGHTS", stem(vecweight), "\n")

# generate another display
boxplot(vecweight, main="Team WEIGHT", ylab="inches")

# generate another
f1 <- fivenum(wt)
cat("five number summary =", f1, "\n")

# compare WEIGHT and HEIGHT
# first produce a scatterplot
plot(vecheight, vecweight, main="Scatterplot of WEIGHT vs HEIGHT")

# now do a linear regression model with the data and store in variable line1
# also print out r
line1 <- lm(vecweight ~ vecheight) 
line1
cat("r=", cor(vecweight,vecheight), "\n")

# plot line of best fit on scatterplot
plot(vecheight, vecweight, main="Scatterplot of WEIGHT vs HEIGHT",
             xlab="inches", ylab="pounds")
abline(line1)

# do scatterplot labeled by position, using first letter
plot(vecweight, vecheight, pch=as.character(vecposition), main="scatterplot of 
              WEIGHT on HEIGHT, Coded with letters")

# do scatterplot labeled by position, using symbol
plot(vecweight, vecheight, pch=as.numeric(vecposition), main="Scatterplot of 
              WEIGHT on HEIGHT, Coded with Symbols", xlab=")

# do residual plot
res1 <- resid(line1)
plot(vecheight, res1, main="Residual Plot-WEIGHT vs HEIGHT",
        xlab="pounds", ylab="residuals")
abline(0,0)		# plot a line with intercept=0 followed by slope=0