2015 H2 Mathematics Prelim Paper 2 Solutions

Qn
1.
(a)

Solutions
Locus of R is a straight line passing through the origin and parallel to AB .

1.
(b)

a and b are parallel

a 2a b b
a 2 a b cos
2

b
1.
(c)

1
2

Using ratio theorem,

1
1
2
ON a b and OM b c
3
2
3
Since O is the midpoint of MN,
ON OM
1
2
1
a b b c
3
3
2
2a 4b 3b 3c

2a 7b 3c 0
2.
(i)

f r 3r 2 3r 1 9r 2 3r 2

f r 9r 2 3r 2
n

r 1

r 1

r 1

r 1

r 1

9 r 2 3 r 2
3
3
n n 1 2n 1 n n 1 2n
2
2
3
2
3n 3n 2n

n 3n 2 3n 2
Hence, a 3 , b 3 and c 2 .
2.
(ii)

S1
S2

3 2 3 1

1
4

1
1
2

4 6 2 6 1 7

qed

2
1
3

7 9 2 9 1 10
3
1
4
S4

10 12 2 12 1 13

S3

Qn
2.
(iii)

Solutions
Conjecture: Sn

n
3n 1

Let P n be the proposition that Sn

Consider P 1 , LHS S1
Therefore P 1 is true.

n
for all n
3n 1

1
RHS .
4

k
.
3k 1
k 1
Want to show P k 1 is true, i.e., Sk 1
.
3 k 1 1
Consider P k 1 ,
Assume P k is true for some k

, i.e., Sk

k 1

1
r 1 f r

S k 1

k
1

3k 1 3 k 1 2 3 k 1 1

k 3k 4 1
3k 1 3k 4

3k 1 k 1
3k 1 3k 4

k 1
3 k 1 1
P k 1 is true.

Since P 1 is true and P k is true P k 1 is true, by mathematical induction,

P n is true for all n

3.
(i)

3
arg 2 2i z
4
3
arg 1 arg z 2 2i
4

arg z 2 2i 0 -----(*)
4

Qn

Solutions
Im

Re

3.
(ii)

Im

max arg
min arg

Re

2
max arg z 3 angle of 0, 2 from 3,0 tan 1
3

2
arg z 3 tan 1
4
3
3.
(iii)

Im

A
B
Re

Area = triangle A + trapezium B

1
1
= 1 2 2 2 4 7
2
2
3

Qn
4.
(i)

Solutions
ln 2 x 1 dx
2
dx
2x 1
1
= x ln 2 x 1 1
dx
2x 1
1
= x ln 2 x 1 x ln 2 x 1 c
2
1

= x ln 2 x 1 x c
2

= x ln 2 x 1 x

4.
(ii)

4.
(iii)

Area of shaded region

1

3/2

ln( 2 x 1)dx ln( 2 x 1)dx

3/4

= x ln( 2 x 1) x ln( 2 x 1)
2

3/4
3/2

+ x ln( 2 x 1) x ln( 2 x 1)
2

Qn

Solutions

3 1 3 1 1
= ( 0 1 0) ln ln
4 2 4 2 2

3 1

+ ln 2 ln 2 ( 0 1 0)
2 2

3
1
= ln 2 square units
4
4

4.
(iv)

ln 5
3
Volume = (3)2 (ln 5) ln 2 x 2dy
ln 2
2

y
ln 5 e 1
3
= (3) (ln 5) ln 2
dy
ln 2
2
2
2

5(i)

= 26.927 = 26.9 cubic units (3s.f.)

In order to use stratified sampling, we need to know the complete composition of
the participants according to strata, e.g. race, gender or age-groups. However, it is
difficult to have this complete information due to absentees, incomplete registration
information, etc. Hence it would be difficult to use a stratified sample.

5(ii)

We assign an index number to each participant when they arrive at the conference.
Let the total number of participants to arrive at the conference be x. Sampling
x
interval =
25 . Pick a random number from 1 to 25 using a random number
0.04 x
generator, then pick every 25th participant thereafter. E.g. If the number 5 was
selected, we sample the participants with numbers 5, 30, 55, until 4% of the
participants are sampled.

6
(ai)

P( A | B) 0.675

P( A B)
P( B)

P( A B) 0.675 P( A B) 0.13
P( A B) 0.675P( A B) 0.08775
P( A B) 0.27

A
0.13

Qn

Solutions

Alternative 1
P A B P A P B P A B

0.62 0.49

P A B
P A B
P A | B

P A B 0.27
Alternative 2
P A ' B
P A' | B
P B 0.4
P B

P A B P B P A | B 0.27
6
(aii)

P( A | B) 0.675
P( A) 1 0.13 0.38 0.49
P( A | B) P( A)
Therefore the two events A and B are not independent
Or

P( A B) 0.27
P( A) P( B) 0.49 0.4 0.196
P( A B) P( A) P( B)
Therefore the two events A and B are not independent
6
(b)

Required Probability
P(CC ) P(CDCC ) P(CDCDCC ) ...
=
P( DCC ) P( DCDCC ) P( DCDCDCC ) ...

p 2 p 2 1 p p 2 1 p ....
2

1 p p 1 p p 1 p p ....
2

1
1
2
p2
1 p p
2
2
1 1 p
1 1 p
2
p p 1 p

2
2p p

1 p p2
2 p
(shown)
9! 10C3 3! 261273600

7(a)

Qn

Solutions

7(b)

[All Universities]=[Any six-person]

[University A&B]
[University B&C]
[University A&B]

Case 1:

University A&B

C6

Case 2: University B&C

Case 3: University A&C

C6

C6

So the answer is C6 C6 C6 C6 805

12

Alternate Method [Not advisable because of too many cases]

Consider number of candidates from each university.
If we have x from A, y from B and z from C, the number of possible ways is
4
Cx 5C y 3Cz
A(4) 1
1
1 2
2
2
3
3
4
B(5) 2
3
4 1
2
3
1
2
1
C(3) 3
2
1 3
2
1
2
1
1
Ans 40 120 60 30 180 180 60 120 15

7(c)

So the total number of possible way is 805.

12
C4 8 C4 4 C4
= 5775
All possible groupings:
3!
Method 1 (Direct Method)
Case 1: Candidates from University A grouped in 2,2,0
(4 from A versus 8 from the rest)
Number of ways = 4C2 8C2 2C2 6C2 2! 1260
Case 2: Candidates from University A grouped in 2,1,1
Number of ways = 4C2 8C2 2C1 6C3 2! 3360
1260 + 3360
So the probability is
= 0.8
5775
Method 2 (Method of Complementation cases where a group has more than 2
candidates from University A)
Case 1: Candidates from University A grouped in 3,1,0
Number of ways = 4C3 8C1 8C4 2! 1120
Case 2: Candidates from University A grouped in 4,0,0
Number of ways = 4C4 8C4 2! 35
1120 + 35
So the probability is 1= 0.8
5775
7

8(a)

2X Y

N 21 2 , 22 4 2 9 2

i.e. 2 X Y

9 X 8Y

N 21 2 , 25 2

N 91 82 ,92 4 2 82 9 2

N 91 82 ,900 2

i.e. 9 X 8Y

P 2 X Y P 9 X 8Y
P 2 X Y 0 P 9 X 8Y 0
2 X Y 2 1 2 0 2 1 2
P

2
25

25 2

9 X 8Y 9 1 82 0 9 1 82
P

900 2
900 2

0 2 1 2
0 9 1 82
P Z
P Z

25 2
900 2

21
82 9 1

P Z 2
P Z

5
30

By symmetry of the standard normal distribution,

21
8 91
So 2
2
5
30
62 121 82 91

142 211
14 2
1

2 21 3

N 10,9

8
(bi)

8
(bii)

Let S be the number of observations with Y 16

Hence S B 100,0.02275

P Y 16 0.97725 = 0.977 (3 sf)

Since n 100 is sufficiently large, np 2.275 5 ,

Po 2.275 approximately

P 100 S 95 P S 5 0.971 (3 s.f.)

17.21 k
, n 3.5
6
n , x lies on the regression line, so
x 0.943n 0.484
17.21 k
0.943 3.5 0.484
6
k 5.50 2 dp

9(i)

9(ii)

6,5.50

9
(iii)

9
(iv)

From the scatter plot, x and n have a curvilinear relationship. Therefore a linear
model is inappropriate even though the product moment correlation coefficient is
relatively high (i.e. 0.915).
The graph of x a bn2 is concave upwards (or increase at increasing rate) similar
to the scatter plot

The graph of x a b ln n is concave downwards (or increase at decreasing rate) .

From GC,
rB 0.8212061688 0.821 3 s.f.

rA 0.9859197289 0.986 3 s.f. which is closer to 1 and hence suggested a

relatively stronger linear relationship between x and ln n as compared to x and n 2 .
Therefore Model (A) is more appropriate.

9
(iv)

Using GC, the regression line of x on ln n is

x 0.638570346 2.869411323ln n

b 2.869411323 2.87
a 0.638570346 0.639
For x 10,
2.869411323ln n 0.638570346 10
ln n 3.262491362
n 26.1

9
(v)
10
(i)

The number of reported cases is at least 1,000 in the 27th month

The value of a represents the estimated number of cases (in hundreds) of the virus
reported in the first month.
Let X 1 , X 2 be the number of customers on Day 1 & Day 2 respectively

Po 20 , X 2 Po 20
Since 20 10,
X 1 N 20, 20 approximately,
Then X1

X2

N 20, 20 approximately.

So X1 X 2

N 0, 40

P X 1 X 2 2 P 2 X 1 X 2 2

10
(ii)

c .c .

P 2.5 X 1 X 2 2.5 0.307

Let Y be the number of customers arrived in n hours
20
So, Y Po n
8
i.e. Y Po 2.5n where n , 0 n 8 .

P Y 2 0.01
P Y 0 P Y 1 P Y 2 0.01
2.5n 0 2.5n 1 2.5n 2
e

0.01
1!
2!
0!
1 2.5n 3.125n 2

0.01
e 2.5 n
When n = 3, P Y 2 0.02026 0.01
2.5 n

When n = 4, P Y 2 0.00277 0.01

So the set of values of n is 4,5,6,7,8 or n

10
(iii)

: 4 n 8

1. The arrival of customers may not be independent, for example, relatives/friends

visit the restaurant together.
2. The mean number of customers per unit time may not be constant throughout the
day. For example, lunch time we may expect more customers.
10

11
(ai)

Let be the population mean time (in seconds) for a school boy in the school to
complete a 4 10 m shuttle run.
H0 : 10.8
H1 : 10.8

11
(aii)

Assumption: Since sample size is small and population variance is unknown, we

need to assume that the time, in seconds, for a school boy to complete a 4 10 m
shuttle run follows a normal distribution.

x 93.96 k

9
9
Sample variance = 1.092
9
2
s 2 1.09 1.3366125
8

Under H0, Test Statistic

X 10.8
t (8)
T
S2
9
x 10.8
Test Statistic value, t =
1.3366125
9
Reject H0 at 5% level of significance,
P T t 0.05
x 10.8
2.306004133
1.3366125
9

93.96 k

10.8
9

2.306004133
1.3366125
9
k 4.758049639
(rejected since k 0)

11

 93.96 k

10.8
9

or
2.306004133
1.3366125
9
k 11.2 (3 sf)

k : k
11
(b)

k 11.2
y

The unbiased estimate for population mean is y

y 517.49

50
50
51749
y
10.3498
5000

The unbiased estimate for population variance is s 2

2
y y
2
s
n 1
122.32 3058
s2
=
2.496326531
49
1225
H0 : 10.8
H1 : 10.8
Under H0, Test Statistic
X 10.8
Z
N 0,1 approximately
3058

1225
50
by Central Limit Theorem since n (=50) is large
From G.C., p-value = 0.0219608 0.0220 (3 s.f.)
Reject H0, % 2.20% and thus 2.20 .

11
(c)

Since n (=50) is large, the mean timing for the boys to complete a 4 10 m shuttle
run will be approximated to a normal distribution by Central Limit Theorem.
Therefore not necessary to assume population follows normal distribution.
2.82
Given n is large, by Central Limit Theorem, X N ,
.
n

P( X 0.8) 0.001

0.8
P Z
0.999
2.8

12

0.8
3.090232308
2.8
n
n 116.98
n 117

13