Week 1. The objectives of the first week include learning how to use the Macintosh to accomplish the following:
Week 2
January 26
February 2
February 9
Week of February 16
Week of February 23
Week of March 2
Week of March 16
Week of March 23
Toss a fair coin and maintain a running count of the number of heads and tails. Suppose that at some time, there are 100 more heads than tails. Is there any tendency for tails to "catch up" in future tosses? The answer depends on what we mean by catching up. We know that future probabilities of a head or a tail are not changed by what happened in the past. Do a simulation of a coin toss. Suppose that heads corresponds to a step up and tails to a step down, and make a plot of the "height" (the number of heads minus the number of tails) as a function of time. You will probably note that unbalanced behavior can persist for a long time. However, it is possible to show that if you wait long enough, the balance will correct itself with probability one. In this sense things do tend to balance out in the long run. But you don't know how long the long run will be.
Suppose that you toss a coin a million times and count the number of times that heads or tails is in the lead. What proportion of the time do you expect heads to have the lead? Do the simulation and see.
Suppose that you toss a coin of 100 times and get 55 heads and 45 tails. If you wait long enough. the balance will correct itself with probability unity. But what if you choose a fixed number of tosses, say a million. What will the number of heads and tails be (given that there was an imbalance after 100 tosses)?
Let us ask similar questions about rolling a die (with six faces). Count the cumulative numbers of each occurrence of each face. What is the probability that however long you throw the die, the six numbers even out at some stage?
Week of March 30
Week of April 6 and 13
Updated 23 April 1998.