THE UNIVERSITY OF HONG KONG

DEPARTMENT OF MATHEMATICS

MATH1013 University Mathematics II

Assignment 2 Solutions

1. (a) We have
√ √ √
x x−1 x x−1 x x+1
lim = lim ( ⋅ √ )
x→1 x − 1 x→1 x−1 x x+1
x3 − 1
= lim √
x→1 (x − 1)(x x + 1)

(x − 1)(x2 + x + 1)
= lim √
x→1 (x − 1)(x x + 1)

x2 + x + 1
= lim √
x→1 x x + 1
3
= .
2

(b) Firstly, by proposition 3.9, we have

2x2 2x2
lim = lim = lim (−2x) = −∞.
x→∞ 1 − x x→∞ −x x→∞

2x2
As lim 3y = 0, we have lim 3 1−x = 0 by proposition 3.11.
y→−∞ x→∞

2x2
Remark. 1 point for showing lim = −∞, and 1 point for computing the whole
x→∞ 1 − x
limit.

(c) When x is close to 1 but not equal to 1, we have f (x) = 2x + 1. Therefore,

lim f (x) = lim (2x + 1) = 3.

x→1 x→1

(d) We have

1 sin2 x cos x − sin2 x

lim (sec x − tan2 x) = lim ( − ) = lim ( ).
x→ 3π
2 x→ 3π
2
cos x cos2 x x→ 3π
2
cos2 x

The limit of cos x − sin2 x is 0 − (−1)2 = −1. The limit of cos2 x is 0. Therefore, the limit is
−1 3π
of the form . When x is close to , cos x − sin2 x < 0 and cos2 x > 0. Thus, the limit
0 2
is −∞ by proposition 3.8(b).

Remark. 1 point for rewriting the function as a single fraction, 1 point for noticing the
L
limit is of the form for some L ≠ 0, and 1 point for the correct limit (with correct
0
explanations).

1
 (e) For any x ∈ (0, 1), we have
1
(x2 + 3) cos + ln x < (12 + 3)(1) + ln x = 4 + ln x.
x2
1
As lim+ (4 + ln x) = 4 + (−∞) = −∞, we have lim+ ((x2 + 3) cos + ln x) = −∞ by propo-
x→0 x→0 x2
sition 3.7(b). It follows that

(x2 + 3) cos x12 + ln x −∞

lim+ = = −∞.
x→0 x+1 0+1

1
Remark. One cannot compute lim+ ((x2 + 3) cos + ln x) by using the extended alge-
x→0 x2
braic rules since the first term has no limit.

2 points for comparing the function with another larger function with limit −∞ and 1
point for the correct limit (with correct explanations).

1
2. (a) Note that −x2 ⩽ x2 sin ⩽ x2 for all x ∈ (0, 2]. As both −x2 and x2 tend to 0 when x
x3
approaches 0, we have
1
lim f (x) = lim+ x2 sin
=0
x→0+ x→0x3
by the sandwich theorem. Therefore, if we define g ∶ [0, 2] → R by

f (x) if x ∈ (0, 2],

g(x) = {
0 if x = 0,

then g is continuous at 0. It is clear that g is continuous at every x ∈ (0, 2]. Therefore, g

is a continuous function.

Remark. 1 point for computing lim+ f (x) and 1 point for defining g(x). It is fine if one
x→0
does not check the continuity of g.

(b) As g is a continuous function defined on [0, 2], it attains its global maximum and global
minimum by the extreme value theorem. These extremum values are not attained at 0
since

g(0) = 0 < sin 1 = g(1),

2
√
2 3 ⎛3 2 ⎞
g(0) = 0 > − ( ) = g .
3π ⎝ 3π ⎠

Therefore, the global maximum and global minimum of g are also the global maximum
and global minimum of f . (More mathematically, if M is the point of global maximum
of g, then M ∈ (0, 2], and hence f (x) = g(x) ⩽ g(M ) = f (M ) for all x ∈ (0, 2]. This shows
f attains the global maximum. Similarly, f attains the global minimum.)

Remark. 1 point for proving 0 is not a global maximum point or a global minimum point
of g and 1 point for proving f attains the global maximum and global minimum.

2
3. Let f (x) = ex − 3x − 4. Note that f (2) ≈ −2.610944 and f (3) ≈ 7.085537. The following table
gives the computation (correct to 6 decimal places).

interval midpoint c f (c)

(2, 3) 2.5 0.682494
(2, 2.5) 2.25 −1.262264
(2.25, 2.5) 2.375 −0.373987
(2.375, 2.5) 2.4375 0.131894
(2.375, 2.4375) 2.40625 −0.126463
(2.40625, 2.4375) 2.421875 0.001340
(2.40625, 2.421875) 2.4140625 −0.062903
(2.4140625, 2.421875) 2.41796875 −0.030867

This shows there is a root in the interval (2.41796875, 2.421875). Since 2.41796875 ≈ 2.42 and
2.421875 ≈ 2.42, a root of f (x) = 0 is 2.42 (correct to 2 decimal places).

Remark. 1 point for showing a correct understanding of the bisection method, 1 point for a
correct computation for at least 4 steps, 1 point for terminating the process at a suitable step
(after showing the root belongs to an interval (a, b) where both a and b are 2.42 correct to 2
decimal places), and 1 point for the correct answer (with correct explanations).