WEBVTT

00:00:00.130 --> 00:00:03.510
A basketball coach has
20 players available.

00:00:03.510 --> 00:00:06.500
Out of them, he needs to choose
five for the starting

00:00:06.500 --> 00:00:10.450
lineup, and seven who would
be sitting on the bench.

00:00:10.450 --> 00:00:14.870
In how many ways can the coach
choose these 5 plus 7 players?

00:00:14.870 --> 00:00:18.660
It is certainly a huge number,
but what exactly is it?

00:00:18.660 --> 00:00:20.800
In this lecture, we will
learn how to answer

00:00:20.800 --> 00:00:22.810
questions of this kind.

00:00:22.810 --> 00:00:26.190
More abstractly, we will develop
methods for counting

00:00:26.190 --> 00:00:29.680
the number of elements of a
given set which is described

00:00:29.680 --> 00:00:31.750
in some implicit way.

00:00:31.750 --> 00:00:34.530
Now, why do we care?

00:00:34.530 --> 00:00:37.340
The reason is that in many
models, the calculation of

00:00:37.340 --> 00:00:39.530
probabilities reduces
to counting.

00:00:39.530 --> 00:00:43.320
Counting the number of elements
of various sets.

00:00:43.320 --> 00:00:46.050
Suppose that we have a
probability model in which the

00:00:46.050 --> 00:00:50.570
sample space, Omega, is finite,
and consists of n

00:00:50.570 --> 00:00:52.500
equally likely elements.

00:00:52.500 --> 00:00:56.450
So each element has
probability 1/n.

00:00:56.450 --> 00:00:59.340
Suppose now that we're
interested in the probability

00:00:59.340 --> 00:01:03.500
of a certain set, A, which
has k elements.

00:01:03.500 --> 00:01:08.750
Since each one of the elements
of A has probability 1/n, and

00:01:08.750 --> 00:01:13.150
since A has k distinct
elements, then by the

00:01:13.150 --> 00:01:17.289
additivity axiom, the
probability of A is equal to k

00:01:17.289 --> 00:01:19.800
times 1 over n.

00:01:19.800 --> 00:01:23.440
Therefore to find the
probability of A, all we have

00:01:23.440 --> 00:01:27.370
to do is to count the number of
elements of Omega and the

00:01:27.370 --> 00:01:30.580
number of elements of A,
and so determine the

00:01:30.580 --> 00:01:33.180
numbers k and n.

00:01:33.180 --> 00:01:36.390
Of course, if a set is described
explicitly through a

00:01:36.390 --> 00:01:40.090
list of its elements, then
counting is trivial.

00:01:40.090 --> 00:01:42.610
But when a set is given
through some abstract

00:01:42.610 --> 00:01:45.950
description, as in our
basketball team example,

00:01:45.950 --> 00:01:48.350
counting can be a challenge.

00:01:48.350 --> 00:01:52.340
In this lecture, we will start
with a powerful tool, the

00:01:52.340 --> 00:01:56.085
basic counting principle, which
allows us to break a

00:01:56.085 --> 00:01:58.350
counting problem into
a sequence of

00:01:58.350 --> 00:02:01.190
simpler counting problems.

00:02:01.190 --> 00:02:05.230
We will then count permutations,
subsets,

00:02:05.230 --> 00:02:08.139
combinations, and partitions.

00:02:08.139 --> 00:02:12.220
We will see shortly what all
of these terms mean.

00:02:12.220 --> 00:02:15.340
In the process we will solve a
number of example problems,

00:02:15.340 --> 00:02:18.600
and we will also derive the
formula for the binomial

00:02:18.600 --> 00:02:21.740
probabilities, the probabilities
that describe

00:02:21.740 --> 00:02:24.430
the number of heads
in a sequence of

00:02:24.430 --> 00:02:26.710
independent coin tosses.

00:02:26.710 --> 00:02:28.329
So, let us get started.