The preceding chapters have introduced ways to calculate the probabilities of random events, based on various assumptions. You may have noticed that many of the problems you've encountered are actually similar, despite the contexts being different. Our approach in the next few chapters will be to classify some of these common types of problems and develop general methods for handling them. Rather than working each problem out as if we'd never seen one like it before, our emphasis will now shift to checking whether the problem is one of these general types.
Before starting on the probability models of the next chapter, it's worth reviewing some useful results for series, and for summing certain series algebraically. We'll be making use of them in the next few chapters, and many, such as the geometric series, have already been used.
Geometric Series:
If
,
then
Binomial Theorem: There are various forms of this theorem.
We'll use the form
In a more general version, when
is not a positive integer,
(We have already defined
when
is not a positive integer.) While the binomial theorem may not look like
a summation result, we'll use it to evaluate series of the form
.
Multinomial Theorem: A generalization of the binomial
theorem is
with the summation over all
with
.
The case
gives the binomial theorem in the form
.
Hypergeometric Identity:
There will not be an infinite number of terms if a and b are positive integers
since the terms become 0 eventually. For example
Sketch of Proof:
Expand each term by the binomial theorem and equate the coefficients of
on each side.
Exponential series:
(Recall, also that
Special series involving integers:
Example: Find
Solution: For
or 1 the term becomes 0, so we can start summing at
.
For
,
we can expand
as
Cancel the
terms and try to re-group the factorial terms as "something choose something".
Then
Factor out
and let
to
get
by the hypergeometric identity.
(Solutions to 5.1 and 5.2 are given in Chapter 10.)
Show
that
(use the binomial theorem).
I have a quarter which turns up heads with probability 0.6, and a fair dime. Both coins are tossed simultaneously and independently until at least one shows heads. Find the probability that both the dime and the quarter show heads at the same time.
Some other summation formulas can be obtained by differentiating the above
equations on both sides. Show that
by starting with the geometric series formula. Assume
.
Players
and
decide to play chess until one of them wins. Assume games are independent with
wins) = .3,
wins) = .25 and
(draw)
= .45 on each game. If the game ends in a draw another game will be played.
Find the probability
wins before
.