Download as pdf or txt
Download as pdf or txt
You are on page 1of 11

Examiners’ commentaries 2016

Examiners’ commentaries 2016


MT2116 Abstract mathematics

Comments on specific questions

Question 1

Parts (a) and (b) of this question build on the material of Chapter 1 of the subject
guide. See section 4.6, on the Pigeonhole Principle, of the subject guide for part (c).

Part (a)

(i) Take any n which is a multiple of 4. By definition, the fact that n is a multiple
of 4 means that there is some natural number m such that n = 4 m. Now we can
write 9n + 30 = 36m + 30 = 6(6m + 5). So by writing k = 6m + 5 we see that we
have 9n + 30 = 6k. And since m is a natural number, k = 6m + 5 is also a natural
number.

So we have shown that there is a natural number k so that 9n + 30 = 6k. This


means that 9n + 30 is a multiple of 6.

(ii) The converse of a statement “if p then q” is the statement “if q then p”. So the
converse of the statement is

If 9n + 30 is a multiple of 6, then n is a multiple of 4.

This statement is false. A statement of type “if q then p” is false if and only if
there is an example where q is true and p is false – i.e., a counterexample. A
counterexample here will be a natural number n so that 9n + 30 is a multiple of 6,
but n is not a multiple of 4. For a counterexample, we can use n = 2:
9n + 30 = 48 = 6 × 8, but n is not a multiple of 4.

Part (b)

We get the following truth tables :

1
MT2116 Abstract mathematics

p q r r ⇒ q p ∧ q (p ∧ q) ⇒ r S
T T T T T T T
T T F T T F F
T F T F F T F
F T T T F T T
T F F T F T T
F T F T F T T
F F T F F T F
F F F T F T T

p q r ¬p ¬q ¬r ¬p ∨ ¬q (¬p ∨ ¬q) ∧ ¬r q ∧ r T
T T T F F F F F T T
T T F F F T F F F F
T F T F T F T F F F
F T T T F F T F T T
T F F F T T T T F T
F T F T F T T T F T
F F T T T F T F F F
F F F T T T T T F T

We find that the two final columns, for S and T , are identical. Hence these statements
are logically equivalent.

Part (c)

Each of the 9 elements of S will be congruent modulo 8 to exactly one of the 8 numbers
from 0 to 7. Since S has 9 elements, there will, by the Pigeonhole Principle, be at least
two elements, say x and y that are congruent to the same number, a, say. But that
means x − y ≡ a − a = 0 (mod 8), and hence 8 divides x − y.

Question 2

Part (a) is on Proof by induction, which is discussed in Chapter 3 of the subject


guide. Part (b) is about functions and their properties which is discussed in Chapter
4 of the subject guide. Part (c) is on equivalence relations, which are discussed in
Chapter 5 of the subject guide.

2
Examiners’ commentaries 2016

Part (a)
n
X
(i) Let P (n) be the statement that (4 k − 3) = n (2 n − 1). The statement P (1) is
k=1
P1
True, since k=1 (4 k − 3) = 4 · 1 − 3 = 1 and 1 · (2 · 1 − 1) = 1. Now suppose
n
X
that P (n) is true, i.e.: (4 k − 3) = n (2 n − 1).
k=1
Then we have
n+1
X n
X
(4 k − 3) = (4 k − 3) + (4 (n + 1) − 3)
k=1 k=1

= (n (2 n − 1)) + (4 n + 1)
= 2 n2 − n + 4 n + 1
= 2 n2 + 3 n + 1
= (n + 1) (2 n + 1)
= (n + 1) (2 (n + 1) − 1).
n+1
X
So we see that (4 k − 3) = (n + 1) (2 (n + 1) − 1), i.e. that P (n + 1) is true. By
k=1
the principle of induction, it follows that P (n) is true for all n ∈ N.
(ii) Let P (n) be the statement that 3n > n3 . The statement P (4) is true, since 34 = 81
and 33 = 27 < 81. Now suppose P (n) is true and n ≥ 4. To show that P (n + 1) is
true, we have to prove that 3n+1 > (n + 1)3 . First, notice that 3n+1 = 3.3n > 3n3 ,
by the inductive assumption that P (n) is true. So it suffices to prove that
3n3 > (n + 1)3 . Now, since n ≥ 4, 3n3 = 3n(n2 ) ≥ 12n2 . Furthermore,
(n + 1)3 = n3 + 3n2 + 3n + 1 ≤ n3 + 3n3 + 3n3 + n3 = 8n3 ,
so we do indeed have 3n+1 > (n + 1)3 , and so P (n + 1) is true. By the principle of
induction, it follows that P (n) is true for all n ≥ 1.

Part (b)

We need to prove f is injective and surjective. For injectivity, we need to prove that for
all z1 , z2 ∈ Z with z1 6= z2 we have f (z1 ) 6= f (z2 ). So, take any z1 , z2 ∈ Z with z1 6= z2 .
If z is odd, then f (z) = z + 3 is even; while if z is even, then f (z) = z + 1 is odd.
So if one of z1 , z2 is even and the other odd, then we certainly have f (z1 ) 6= f (z2 ).
This means we only have to consider the case that z1 , z2 are both even or both odd.
If z1 , z2 are both even, then f (z1 ) = z1 + 1 and f (z2 ) = z2 + 1. Since z1 6= z2 , we have
z1 + 1 6= z2 + 1, hence f (z1 ) 6= f (z2 ).
And if z1 , z2 are both odd, then f (z1 ) = z1 + 3 and f (z2 ) = z2 + 3, and exactly the same
argument leads to f (z1 ) 6= f (z2 ). To establish surjectivity, we need to prove that for all
y ∈ Z, there is an x ∈ Z so that f (x) = y. Take any y ∈ Z. We distinguish two cases :

3
MT2116 Abstract mathematics

If y is even, then set x = y − 3. Then x ∈ Z and x is odd, hence


f (x) = x + 3 = (y − 3) + 3 = y.

If y is odd, then set x = y − 1. Then x ∈ Z and x is even, hence


f (x) = x + 1 = (y − 1) + 1 = y.

Since an integer is either even or odd, this completes the proof.

Part (c)

Recall what it means to say that a relation R on a set A is an equivalence relation. It


means three things:

• R is reflexive: for all x ∈ A, x R x.


• R is symmetric: for all x, y ∈ A, x R y implies y R x.
• R is transitive: for all x, y, z ∈ A, whenever x R y and y R z, we also have x R z;
that is, (x R y) ∧ (y R z) =⇒ xRz.
R is reflexive because for any a ∈ Z, a + 3a = 4a is a multiple of 4.
R is symmetric. For, suppose x R y. Then x + 3y is a multiple of 4: say, x + 3y = 4k
where k ∈ Z. So, x = 4k − 3y. Then,
y + 3x = y + 3(4k − 3y) = −8y + 12k,
and this is a multiple of 4.
R is transitive. Suppose x R y and y R z. Then x + 3y and y + 3z are multiples of 4. So,
(x + 3y) + (y + 3z) is a multiple of 4: say it’s equal to 4r where r ∈ Z. But
(x + 3y) + (y + 3z) = x + 3z + 4y,
so x + 3z = 4r − 4y = 4(r − y) is a multiple of 4 and hence x R z.

Question 3

See Chapters 6 and 7 of the subject guide for the related background material for
this question.

Part (a)

Given that 2p − 1 is a prime, the divisors of n = 2p−1 (2p − 1) are the numbers 2i for
i = 0, 1, . . . , p − 1 and (2p − 1)2i for i = 0, 1, . . . , p − 1. Therefore
p−1 p−1 p−1
X X X
i i p p
σ(n) = 2 + 2 (2 − 1) = 2 2i = 2p (2p − 1) = 2n.
i=0 i=0 i=0

So n is perfect.

4
Examiners’ commentaries 2016

Part (b)

By the Euclidean Algorithm, we have


357 = 290 + 67
290 = 4 × 67 + 22
67 = 3 × 22 + 1,
so 290 and 357 are coprime, from which it follows that 290 is invertible. And,
1 = 67 − 3 × 22
= 67 − 3(290 − 4 × 67)
= 13 × 67 − 3 × 290
= 13(357 − 290) − 3 × 290
= 13 × 357 − 16 × 290.
The fact that 13 × 357 − 16 × 290 = 1 means that, modulo 357, −16 × 290 ≡ 1. So, in
Z357 , (290)−1 = 16 = 341.

Part (c)

The fact that 2a ≡ 2b (mod 2m) means 2m|(2a − 2b). Thus, for some integer k,
2a − 2b ≡ k(2m), so a − b = km is a multiple of m, and a ≡ b (mod m). Finally, we have
2(2) ≡ 2(5) (mod 6), but 2 is not congruent to 5 mod 6.

Question 4

See Chapter 8 of the subject guide for the relevant background material for all parts
of this question.

Part (a)

Suppose the number is x. Then 1000x = 301.23123 and


30093
1000x − x = 999x = 301.23 − 0.30 = 300.93 = ,
100
so
30093
x= .
99900

Part (b)

We have
|z + iz|2 = (z + iz)(z + iz) = (z + iz)(z − iz) = zz + iz 2 − iz 2 + zz = 2|z|2 + i(z 2 − z 2 ).

5
MT2116 Abstract mathematics

Now, for any w ∈ C, w − w = 2i Im(w), so (with w = z 2 , and noting that z 2 = (z)2 , the
expression equals
2|z|2 − i(2i Im(z 2 )) = 2|z|2 + Im(z)2 .

Part (c)

Denote the polynomial by p(z). Then


p(i) = i4 − 4i3 + 6i2 − 4i + 5 = 1 + 4i − 6 − 4i + 5 = 0,
so z = i is a solution of the equation. It follows, since complex roots of polynomials with
real coefficients occur in conjugate pairs, that −i is also a solution. Hence
(z − i)(z + i) = z 2 + 1 is a factor of p(z). We therefore have
p(z) = (z 2 + 1)(az 2 + bz + c).
Evidently, a = 1 and c = 5. To find b, we can examine the terms in z 3 . The polynomial
p(z) has −4z 3 , whereas expanding the product gives bz 3 , so we must have b = −4.
Therefore,
z 4 − 4z 3 + 6z 2 − 4z + 5 = (z 2 + 1)(z 2 − 4z + 5),
and the other two solutions are the solutions of z 2 − 4z + 5 = 0, which are

4 ± −4
= 2 ± i.
2

Question 5

See Chapter 10 of the subject guide for the background material to parts (a) and (c)
of the question. See Chapter 11 f the subject guide for the relevant background
material for part (b).

Part (a)

You must get this definition right, otherwise this whole part of the question cannot be
answered correctly. We have: xn → L as n → ∞ if, for all  > 0, there is N such that
|xn − L| <  for all n > N . We note that
|xn − xm | = |xn − L + L − xm | = |xn − L − (xm − L)| ≤ |xn − L| + |xm − L|.
Since (xn ) converges to L, for ε > 0, we can find N such that for n > N , we have
|xn − L| < ε/2. Hence, for n, m > N , we have
|xn − xm | ≤ |xn − L| + |xm − L| < ε/2 + ε/2 = ε.

Suppose that (yn ) has a convergent subsequence (ykn ). Let xn = ykn . Then (xn )
converges to some L. So, by the above, there is N such that for n, m > N , we have
|xn − xm | < 1/100. On the other hand, by our assumption, for any n, m, we have
|xn − xm | = |ykn − ykm | > 1/100,
and this is a contradiction.

6
Examiners’ commentaries 2016

Part (b)

We first notice that g(2016) = (2016 − 2016)f (2016) = 0. To prove that g is continuous
at 2016, we need to show that for every  > 0 there exists δ > 0 such that for every x
such that |x − 2016| < δ, we have |g(x) − g(0)| < . We look at |g(x) − g(0)|:

|g(x) − g(0)| = |g(x)| = |(x − 2016)f (x)| = |x − 2016||f (x)|.

We know that |x − 2016| < δ and that |f (x)| < 2. So, we set δ = ε/2 and obtain:

|g(x) − g(0)| = |x − 2016||f (x)| < δ · 2 = ε.

Part (c)

We use the Algebra of Limits.


√ √
√ √ ( n2 + n)2 − ( n2 − n)2
lim ( n2 +n− n2 − n) = lim √ √
n→∞ n→∞ n2 + n + n2 − n
(n2 + n) − (n2 − n)
= lim q q 
n→∞
n 1 + n1 + 1 − n1

2
= lim q q
n→∞ 1 1
1+ n
+ 1− n

2
=√ √
1+0+ 1−0
= 1.

Question 6

Part (a) is on the limits of functions which is discussed in Chapter 11 of the subject
guide. Part (b) looks at the concepts of suprema and infima, which are covered in
Chapter 9 of the subject guide. Part (c) draws on concepts of Chapter 10 of the
subject guide.

Part (a)

f (x) tends to L as x tends to a means that for any  > 0 there is δ > 0 such that if
0 < |x − a| < δ, then |f (x) − L| < .
“Using this formal definition” means showing that for any  > 0 there is some δ > 0
(which will depend on  and which we will produce) such that if 0 < |x − 2| < δ then
|f (x) − 1| < . (It is not appropriate to use results on the ‘algebra of limits’ here
because the question specifically asks us to use the formal definition.) Now,

|f (x) − 1| = (x − 1)2 − 1 = x2 − 2x = |x||x − 2|.

7
MT2116 Abstract mathematics

We’re thinking of x as being close to 2. Let’s suppose, in particular, that |x − 2| < 1, so


1 < x < 3. Then we have |x| < 3. In this case,

|x||x − 2| < 3|x − 2|

and we can make this less than  > 0 if, additionally, |x − 2| < /3. So, if both
|x − 2| < 1 and |x − 2| < /3 then we will have |f (x) − 1| < . Thus, if we let
δ = min(1, /3), it will be the case that |x − 2| < δ implies |f (x) − 1| <  and we are
done.

Part (b)

Suppose that τA and τB are the infima of A and B, respectively. Then, for all a ∈ A and
for all b ∈ B, we have a ≥ τA and b ≥ τB . In particular, therefore, if τ = min{τA , τB }
then for all a ∈ A and b ∈ B, a ≥ τ , b ≥ τ . Any c ∈ C belongs to either A or B and
hence satisfies c ≥ τ . So, C is bounded below by τ . It follows that C has a greatest
lower bound, τC and that τC ≥ τ . (It does not yet follow that τC = τ : for that, we need
to do more work!) We might as well suppose that τA ≤ τB (for, if not, a very similar
argument can apply). In other words, we may assume ‘without loss of generality’ that
τA ≤ τB . So this means τ = τA . Now, since τA is the greatest lower bound for A, if
τ 0 > τ then there is some a ∈ A such that a < τ 0 . But since C = A ∪ B, this a belongs
to C. So, we see that: for any τ 0 > τ there is some c ∈ C with c < τ 0 . This shows that τ
is the greatest lower bound for C. That is, τC = τ , as required.

Part (c)

Certainly, we have x1 = 1 ≤ 5/4. Suppose, as suggested, that n ≥ 2 and that


xn−1 ≤ 5/4. We want to show that this means also that xn ≤ 5/4. Now,
1 1 5
xn = (2xn−1 + 5) ≤ (2(5/4) + 5) = ,
6 6 4
which is what we need. So, yes, we do have xn ≤ 5/4 for all n. Having shown that the
sequence is bounded above, we can prove that it converges if we can show that it is an
increasing sequence. To show the sequence is increasing, we show that xn − xn−1 ≥ 0.
We have
1 5 2 5 25
xn − xn−1 = (2xn−1 + 5) − xn−1 = − xn−1 ≥ − = 0.
6 6 3 6 34
So, as mentioned, because the sequence is bounded above and increasing, it must
converge.
If xn → L then xn−1 → L too. But that means, by the algebra of limits, that
1 1
xn = (2xn−1 + 5) → (2L + 5),
6 6
so we must have
1
L = (2L + 5)
6
and so L = 5/4.

8
Examiners’ commentaries 2016

Question 7

See Chapters 12 and 13 of the subject guide for the relevant background material to
this question.

Part (a)

See Chapter 12 of the subject guide for relevant background material. Let (G, ?) be a
group with identity e. The inverse of a ∈ G is an element a−1 of G satisfying
a ? a−1 = a−1 ? a = e. But this means that also a−1 ? a = a ? a−1 = e. We know that the
inverse is unique, hence (a−1 )−1 = a. For a, b ∈ G, we notice that

(a ? b−1 ) ? (b ? a−1 ) = a ? (b−1 ? b) ? a−1 = a ? e ? a−1 = a ? a−1 = e.

and
(b ? a−1 ) ? (a ? b−1 ) = b ? (a−1 ? a) ? b−1 = b ? e ? b−1 = b ? b−1 = e.
Again, since the inverse is unique, we have that (a ? b−1 )−1 = b ? a−1 .

Part (b)

Recall, from Chapter 12 of the subject guide, that (S, ∗) is a group if and only if the
following properties hold:

• ∀x, y ∈ S, x ∗ y ∈ S [closure property]

• ∀x, y, z ∈ S, (x ∗ y) ∗ z = x ∗ (y ∗ z) [associativity property]

• ∃e ∈ S such that ∀x ∈ S, e ∗ x = x ∗ e = x [identity property]

• ∀x ∈ S, ∃x−1 ∈ S such that x ∗ x−1 = x−1 ∗ x = e [inverse property]


We first observe that
      
−1 −1
 a b  c d   ac ad + bc
=  =  ac ad + bc  ∈ G.
0 a−1 0 c−1 0 c−1 a−1 0 (ac)−1

Here we used the fact that a, c 6= 0 and, hence, part (a) applies to the multiplicative
group Z∗5 .
 
1 0 
Secondly, clearly  ∈ G by taking a = 1 and b = 0.
0 1
 
a b 
Finally, we need to find an inverse of  ∈ G. In the above, we set
−1
0 a
ac = 1 = (ac) and ad + bc = 0. Again, since a 6= 0, in the multiplicative group Z∗5 ,
−1 −1

we get c = a−1 . Hence, we have ad + b(a−1 )−1 = 0, from which we obtain that ad = −ba,
or da = −ba because Z∗5 is abelian and (a−1 )−1 = a. By multiplying by a−1 , we get

9
MT2116 Abstract mathematics
   
−1 −1
a −b   a −b 
d = −b. So,  = ∈ G is the inverse. The order of G is 20
0 a 0 (a−1 )−1
because there are 4 ways to choose a 6= 0 and 5 ways to choose b.

Part (c)
 
1 0 
H is non-empty because  ∈ H. We also see that
0 1
      
−1
 a 0   c 0  =  ac a0 + 0c  =  ac 0 
∈H
0 a−1 0 c−1 0 (ac)−1 0 (ac)−1

and
 −1  
−1
 a 0  a −0  ∈ H.
=
0 a−1 0 −1 −1
(a )

Here we used part (b) and the fact that a 6= 0 and c 6= 0.

Question 8

See Chapters 12, 13 and 14 of the subject guide for the relevant background material
to this question.

Part (a)

First, we show closure. For any x, y ∈ G, xgy ∈ G since G is a group, so x ∗ y ∈ G.


Next, we show associativity. Let x, y, z ∈ G. Then

(x ∗ y) ∗ z = (xgy) ∗ z = (xgy)gz

and
x ∗ (y ∗ z) = x ∗ (ygz) = xg(ygz).

But because multiplication in G is associative, these two expressions are equal to each
other.
If x ∈ G then x ∗ g −1 = xgg −1 = x and g −1 ∗ x = g −1 gx = x, so e = g −1 is an identity
element for the operation.
Suppose x ∈ G. To determine an inverse for x, we need b such that
x ∗ b = b ∗ x = e = g −1 . That is, we need xgb = bgx = g −1 . This is satisfied by
b = g −1 x−1 g −1 , and so every element of G has an inverse.

10
Examiners’ commentaries 2016

Part (b)

We need to show that λx is injective and surjective. Now,

λx (g) = λx (h) =⇒ xg = xh =⇒ x−1 xg = x−1 xh =⇒ g = h,

so it is is injective. Furthermore, for any g ∈ G, we have g = λx (x−1 g) since

λx (x−1 g) = x(x−1 g) = (xx−1 )g = eg = g.

This shows that λx is surjective.


Next, we show that, for any g,

λx (λy (g)) = λxy (g).

We have
λx (λy (g)) = λx (yg) = x(yg) = (xy)g = λxy (g).

To say that a mapping θ : G → H from a group (G, ?) to a group (H, ·) is a


homomorphism means that θ(x ? y) = θ(x) · θ(y) for all x, y ∈ G. (See Chapter 14 of the
Subject Guide.)
Now, in the present context, ? is the operation in G, written multiplicatively, and · is
composition of functions, so this means precisely

λxy = λx λy .

We’ve just shown that for all g ∈ G,

λxy (g) = λx (λy (g)) = (λx λy )(g),

so we see that θ is indeed a homomorphism. Now, to show that θ is injective, we must


show that if θ(x) = θ(y) then x = y. That is, we must show that if the two functions λx
and λy are equal, then x = y. Certainly, λx = λy implies λx (e) = λy (e), which means
x = y.
An isomorphism between groups is a bijective homomorphism. We have not shown that
θ is bijective, only that it is injective. But if H = im θ = {θ(x) : x ∈ G}, then H is a
subgroup of S and θ : G → H will be both injective and surjective. It then follows that
θ is an isomorphism between G and H, a subgroup of S.

11

You might also like