Permcomb

PERMUTATIONS AND COMBINATIONS
S. VISWANATH
1. General Principles of Counting

Example 1. (a) Find the total number of three letter words. Here, by a
word, we mean any string of three alphabets. (b) Now, find the number of
such words which contain only vowels or only consonants.
Since there are 26 choices for each of the three letters in the word, the total
number in part (a) is 263 . In part (b), we subdivide this into two problems:
counting those words which contain only vowels (53 ) and those which contain
only consonants (213 ). The answer is then the sum 53 + 213 .
Example 2. Find the number of 8 digit numbers formed only using the
digits 4, 5, 6 which are (a) even (b) divisible by 5 (c) divisible by 3.
(a) For such a number to be even, the units digit has to be even, hence
4 or 6. Thus there are two choices for the units digit, and 3 choices for
each of the remaining 7 digits. Hence the total number of such numbers is
37 × 2. (b) Similar, with only one choice for units digit (namely 5). Thus,
here the count is 37 . (c) This is more complicated since all digits play a
role in determining if a number is divisible by 3. We need the sum of all 8
digits to be a multiple of 3. We proceed as follows: first choose the first 7
digits arbitrarily (37 ways of doing this). The units digit is now uniquely
determined by the requirement that the sum of digits is divisible by 3; this
is because the set of allowed digits {4, 5, 6} leave different remainders on
division by 3, and exhaust all remainders. Thus, here again the count is
37 .
Problem 1. Redo part (c) of example 2 under the assumption that the
allowed digits are now 1, 2, 3, 5, 7, 9.
Problem 2. Pose your own variations of problem 1, and explore ways of
solving them.
We used two general principles in the above examples:
(1) The sum principle: If the set of objects to be counted can be sepa-
rated into two disjoint subsets, then the total number of objects is
the sum of the numbers of objects in the two subsets.
(2) The product principle: If an object A can be chosen in m ways, and
once A is chosen, an object B can be chosen in n ways, then the
number of ways of choosing the ordered pair (A, B), i.e., first A,
then B, is mn.
1
2 S. VISWANATH
2. Permutations
Given a set S of objects, a permutation of S is a way of arranging these
objects in a line. If S has n elements, a permutation can be formally defined
as a function f : {1, 2, · · · , n} → S such that f is injective (i.e., one-to-one).
Problem 3. (a) Convince yourself that this formal definition is the same as
the more intuitive notion of a permutation as an arrangement. (b) Suppose
A, B are finite sets with the same number of elements. Prove that a function
f : A → B is one-to-one if and only if it is onto. In other words, f is injective
if and only if f is bijective.
The number of permutations of n distinct objects is n(n−1)(n−2) · · · (2)(1).
This is defined to be the factorial of n and denoted n!. It is often convenient
to take the n distinct objects to simply be the numbers 1, 2, · · · , n. More
generally, if r is a number with 0 ≤ r ≤ n, then a permutation of n distinct
objects taken r at a time is an arrangement of any r out of the n objects in
a line.
Problem 4. Check that this is the same thing as a function f : {1, 2, · · · , r} →
S which is injective.
n!
The number of permutations of n objects taken r at a time is (n−r)! , and
is denoted nPr .
Example 3. Find the number of ways of placing 8 rooks on a chessboard
such that no two of the rooks can attack each other.
A chessboard is an 8 × 8 grid of squares, and a rook (or elephant) can
move horizontally or vertically on the board to attack other pieces. So, we
need to find the number of ways of choosing 8 squares on the chessboard
such that there is exactly one square in each row and exactly one square
in each column. Suppose we have one such configuration. We can encode
this in the following table, where we have noted down the coordinates (row,
column) of each rook.
Row 1 2 3 4 5 6 7 8
Col c1 c2 c3 c4 c5 c6 c7 c8
Now observe that c1 , c2 , · · · , c8 form a permutation of 1, 2, · · · , 8. Conversely
given such ci ’s, we obtain a non-attacking rook configuration. Thus, the
number of required configurations is equal to the number of permutations
of 8, which is 8!.
Problem 5. More generally, let k be an integer between 0 and 8. Find the
number of non-attacking configurations of k rooks on the chessboard.
Problem 6. In the previous problem, we are assuming that the rooks are
all identical. Redo the above problem under the assumption that the k rooks
are of k different colours (i.e., distinguishable).
3
3. Combinations
The number of ways of choosing r objects from n distinct objects is de-
noted n Cr . We have the formula:
n n!
Cr = .
r!( (n − r)!
Example 4. Let r be a positive integer. Show that the product of r con-
secutive natural numbers is always divisible by r!.
Let us denote the r consecutive numbers (in descending order) as n, n −
1, n − 2, · · · , n − r + 1. Then their product n(n − 1) · · · (n − r + 1) is exactly
nP
r
n!/(n − r)!, which is just nPr . So, we need to show that is an integer.
r!
n!
But observe that this is just r! (n−r)! = n Cr , which is clearly an integer.
Example 5. Consider all words (meaningful or not) that use only the letters
a, b. Find the number of such words of length n which have r occurrences
of a and (n − r) occurrences of b.
Any such word is specified uniquely once we know the positions in which
the letter a occurs. Thus, we need to pick r positions from the total available
n positions. The number of possibilities is thus n Cr .
The previous example gives a visual proof of the binomial theorem. Let
us consider two variables a, b and fully expand the product (a + b)n = (a +
b)(a + b) · · · (a + b). While expanding, let us avoid using the commutativity
property, i.e., we don’t allow ourselves to replace ab by ba. If we do this for
small values of n, say n = 2, 3, we get:
(a + b)(a + b) = aa + (ab + ba) + bb.
(a + b)(a + b)(a + b) = aaa + (aab + aba + baa) + (abb + bab + bba) + bbb.
For general n, the corresponding right hand side will be a sum of all words
of length n in a, b. Now, allowing ourselves commutativity of a, b again, we
observe that the term ar bn−r will occur in the expansion as many times as
there are words of length n in a, b which contain r occurrences of a (and
n − r occurrences of b). By the preceding example, this is just n Cr . We thus
obtain the binomial theorem:
Xn
n n
(a + b) = Cr ar bn−r .
r=0
Problem 7. Carry out a similar analysis with three variables a, b, c. More

precisely, given non-negative integers i, j, k with i + j + k = n deduce that
the coefficient of ai bj ck in (a + b + c)n is equal to the number of words of
length n in a, b, c in which a, b, c occur i, j, k times respectively. Further
n!
prove that this number equals .
i! j! k!
4 S. VISWANATH
Problem 8. Generalise the above problem to k variables, for any k ≥ 2.

The resulting expansion is called the multinomial theorem.
We recall that the key identity satisfied by the binomial coefficients is:
n+1
Cr+1 = n Cr + n Cr+1 .
This is what is used to construct each row of the Pascal’s triangle from
the preceding row. We now show how this identity can be proved without
explicitly using the formula in terms of factorials. The LHS is the number of
ways of choosing r + 1 numbers from the numbers 1, 2, · · · , n + 1. Each such
choice either contains the number n + 1 or does not contain it. Let us count
the number of choices of each type. If we are not allowed to choose n + 1,
we must then choose all our r + 1 numbers from within the set 1, 2, · · · , n.
The number of such ways is thus n Cr+1 . On the other hand, if we have to
pick the number n + 1, we only need to pick the remaining r numbers from
amongst the numbers 1, 2, · · · , n. The number of choices here is n Cr . The
sum principle then establishes the required equality.
4. Combinations with repetitions

Problem 9. There are 4 types of flowering plants in a garden: Jasmine,
Lotus, Rose, Sunflower. Assume that there are a large number of flowers
of each type. We want to pluck a total of 10 flowers and put them in our
basket. In how many different ways can this be done ?
Problem 10. Find the number of solutions to the equation:
x + y + z + w = 10,
where x, y, z, w must be non-negative integers.

Permcomb

Uploaded by

Copyright:

Available Formats

Permcomb

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Permcomb

Uploaded by

Copyright:

Available Formats

PERMUTATIONS AND COMBINATIONS

1. General Principles of Counting

Problem 7. Carry out a similar analysis with three variables a, b, c. More

Problem 8. Generalise the above problem to k variables, for any k ≥ 2.

4. Combinations with repetitions

You might also like