Supplementary Exercise 2.67 of IPS7e
------------------------------------

Negative association between the number flu cases and the amount of ice cream
sold across the weeks of a year. 

This is a simple example of an association derived from common response
situation. This means that a third variable, here the season or weather 
corresponding to the weeks, affects both the variables of interest (number
of flu cases and ice cream sales). The season/weather is clearly a 
confounding variable, and we would think that there is no real
relation between the number of flu cases and the ice cream sales.

Let
  y = ice cream sales
  x = flu cases
  z = season/weather

We could then draw a diagram with the arrows:
- arrow from z to x,
- arrow from z to y,
- but no arrow between y and x.

---

Added note: There is almost no end to examples of this type. Here is
another example (from the IPS textbook) - a positive association between
the number of firefighters at a fire and the damage the fire does.
Explain why we should not conclude from this information that sending 
more firefighters to a fire causes more damage.