Combinatorics

Combinatorics is the branch of maths concerned with counting. A permutation is an arrangement with regard to order, whereas in a combination the order does not matter. For example, the letters AB and BA are two different permutations, but the same combination of letters.

Permutations

Permutations of \(n\) objects

If we have 3 distinct objects, labelled \(A\), \(B\) and \(C\), how many permutations are there?

There are 6 permutations: \(ABC, ACB, BAC, BCA, CAB\) and \(CBA\).

More generally, suppose we have \(n\) different objects. In how many ways can \(n\) different objects be arranged? Each arrangement (i.e. respecting different orders) is a permutation. To obtain a general expression for the number of permutations, imagine there are \(n\) empty boxes and \(n\) items labelled \(1\), \(2\), …, \(n\), to put in the \(n\) boxes. One way of putting the items in the boxes would be:

Box 1

Box 2

.

.

.

Box \(n\)

Item 1

Item 2

.

.

.

Item n

So, let’s start with the \(n\) empty boxes and think about ways in which we can fill them.

  • There are \(n\) ways of filling the first box (we can pick item 1, item 2, …, or item \(n\)).

  • Once we have filled the first box, we have \((n-1)\) objects left and \((n-1)\) empty boxes left.

  • There are \((n-1)\) ways of filling the second box, leaving \((n-2)\) objects to choose from.

  • There are \((n-2)\) ways of filling the third box, leaving \((n-3)\) objects.

  • There is just one way of filling the (last) \(n\)th box.

For each of the \(n\) ways of filling the first slot, there are \(n-1\) ways of filling the second slot, i.e. there are \(n(n-1)\) ways of filling the first two slots. For each of these \(n(n-1)\) ways, there are \(n-2\) ways to fill the third slot, etc. So there are \(n\times(n-1)\times(n-2)\times...\times3\times2\times1\) ways of arranging the \(n\) objects, i.e.the number of permutations of \(n\) objects is equal to

\[ n \times (n-1) \times \ldots \times 3 \times 2 \times 1 = n! \]

We denote this expression by \(n!\) (pronounced \(n\) factorial). As a matter of convention we set \(0!=1\).

Permutations of a subset of \(x\) objects chosen from \(n\)

Now consider, how many possible permutations are there of \(x\) objects if \(x\) objects are selected from the total \(n\) possible? Once again, we can think of the number of ways that we can fill \(x\) boxes, as in the diagram below.

Box 1

Box 2

.

.

.

Box \(x\)

?

?

.

.

.

?

So, let’s start with the \(x\) empty boxes and think about ways in which we can fill them with the \(n\) items.

  • There are \(n\) ways of filling the first box, leaving \((n-1)\) objects and \((x-1)\) boxes.

  • There are \((n-1)\) ways of filling the second box, leaving \((n-2)\) objects to choose from.

  • There are \((n-2)\) ways of filling the third box, leaving \((n-3)\) objects.

  • There are \((n-x+1)\) ways of filling the (last) \(x\)th box.

So there are \(n \times (n-1) \ldots \times (n-x+1)\) ways of arranging (or permuting) \(x\) objects from \(n\).

We often write \(_n P_x\) or \(^n P _x\) to denote the number of permutations of \(x\) objects from \(n\) different objects.

Including \((n-x)(n-x-1)...\times 3 \times 2 \times 1\) in the numerator and denominator, we can express the number permutations of \(x\) objects from \(n\) as:

\[ ^n P _x = \frac{n(n-1)(n-2)...(n-x+1)(n-x)(n-x-1)...\times 3 \times 2 \times 1}{(n-x)(n-x-1)...\times 3 \times 2 \times 1} = \frac{n!}{(n-x)!} \]

We can think of the number of permutations of \(x\) objects chosen from \(n\) as the number of permutations of \(n\) objects, divided by the number of of permutations of the \((n-x)\) objects that we don’t choose.

Combinations

Combinations of \(x\) objects chosen from \(n\)

Suppose now we select \(x\) objects chosen from \(n\), but we are not concerned with the order of the \(x\) objects.

Let \(k\) be the number of ways of choosing \(x\) from \(n\) objects without regard to order, i.e. the number of combinations of \(x\) objects from \(n\) objects. From above, we know that for each of these combinations (of \(x\) objects) there are \(x!\) permutations. Hence

\[ k = \frac{n!}{x! (n-x)!} \]

Intuitively, there are more permutations than combinations, since for every combination there are several permutations:

\[ \frac{n!}{(n-x)!} > \frac{n!}{x! (n-x)!} \]

We often write \(^n C_x\) or the binomial coefficient \(\begin{pmatrix}n \\ x \end{pmatrix}\) to denote the number of combinations of \(x\) objects from \(n\),

\[\begin{split} ^n C_x = \begin{pmatrix}n \\ x \end{pmatrix} = \frac{n!}{x! (n-x)!}. \end{split}\]