setwd('C:/local/classes/M588')

##17
data <- read.csv("pollen.csv", header = TRUE)     
attach(data)
 names(data)
Y <- REMOVED[BEE == 'QUEEN']
X <- DURATION[BEE == 'QUEEN']

lm1 <- lm(Y ~ X)
summary(lm1)
plot(X, Y)
  abline(lm1)
lm2 <- lm(Y[X<33] ~ X[X<33])
summary(lm2)
  abline(lm2, lty = 2)
 # the two outliers affected the fit greatly. Later, we will use influence diagnostics to detect this
   par(mfrow=c(2,2))
   plot(lm1)
 detach(data)

##26
data <- read.csv("male-birth.csv", header = TRUE)     
attach(data)
 names(data)

lm3 <- lm(denmark ~ year) 
summary(lm3)
plot(year, denmark)
  abline(lm3)

plot(year[!is.na(denmark)], lm3$res)