Appliances <- read.csv("Appliances.csv")

Appliances.ts <- ts(Appliances$Shipments, start=c(1985,1), end=c(1989,4), freq=4)
plot(Appliances.ts, main="U.S. Shipments of Household Appliances 
From 1985-1989
(Quarterly)", xlab="Time", ylab="Shipments in Millions", ylim=c(3944, 4900), bty="l")

# The level seems to be around 4400. There does appear to be seasonality. Q1 and Q4 seem to have less shipments than Q2 and Q3.


# Changing the scale might make it easier to interpret.

plot(Appliances.ts, main="U.S. Shipments of Household Appliances 
     From 1985-1989
     (Quarterly)", log="y", xlab="Time", ylab="Shipments in Millions (log scale)", ylim=c(3944, 4900), bty="l")

# I don’t really think they look very different.

Appliances["logShipments"] <- log(Appliances$Shipments)
plot(Appliances$logShipments, main="U.S.Shipments of Household Appliances 
From 1985-1989
Quarterly", ylab="Shipments in Millions (Log Scale)", type="l", bty="l")

yearlyShipments <- aggregate(Appliances.ts, nfrequency=1, FUN=sum)
plot(yearlyShipments, main="U.S. Household Appliance Shipments From 1985-1989
     Aggregated Yearly", ylab="Shipments in Millions", bty="l")