HWA CHONG INSTITUTION

MATHEMATICS DEPARTMENT PAGE 1

2011 GCE A Level H2 Maths Solution Paper 2
SECTION A (PURE MATHEMATICS)

1(i)


The shaded region above represents the region in which
z

lies.

Note: Unless required in the
question, it would be sufficient to
just indicate the centre and radius
of the circle in such a locus
drawing.
1(ii)


Minimum z is given by OA

OC AC =
2 2
2 5 3 29 3 = + =

Maximum z is given by OB
OA AB = +
( )
29 3 6 29 3 = + = +














1(iii)


Remark 1: The length DP in
this case, is always the same
regardless of which coordinates
of P we are taking. This is
because it happens that the line
through CD is perpendicular to
the line representing
( )
1
arg
4
z t = . This can be easily
verified by finding the gradient of
the line through CD.

Re
Im
3





Re
Im
C

Re
Im
C
P
P
D

HWA CHONG INSTITUTION
MATHEMATICS DEPARTMENT PAGE 2

Combining the restriction given by ( )
1
0 arg
4
z t s s , the
resulting locus of z

is given as the shaded region above
(see diagram).

The point D

represents the complex number 6 i + .

The maximum value of 6 i z

is given by DP (there
are two possible such positions that give this maximum
value).

Observe that ( ) 2, 2

and ( ) 5, 5 both lie on the circle
which has Cartesian equation ( ) ( )
2 2
2
2 5 3 x y + = .
Thus, these two points are the possible positions of P .
This means the maximum value required is DP
( ) ( )
2 2
6 2 1 2 = + 17 = .











Remark 2: This part can also be
solved by changing the equations
of the circle and the half-line to
Cartesian form and do
simultaneous equations solving.
The equation of the circle is
already given while the equation
of the half-line is y x = .
2(i) From the questions description, the box will have base
area PS SR and height x .
Thus the volume is given by
( )( ) 2 2 2 V n x n x x =

( )
2 2
2 2 2 x n nx xn x = +
2 2 3
2 6 4 n x nx x = + . (Shown!)

2(ii)
Find
2 2
d
2 12 12
d
V
n nx x
x
= + and let
d
0
d
V
x
=
( ) ( )
( )
2
2
6 6 4 6
12
n n n
x

=
( )
6 12
1 3
12 2 6
n
x n

| |
= =
|
|
\ .

From the diagram (for example, from the side of AD),
we deduce that 2x n < , and thus
1 3
2 6
x n
| |
=
|
|
\ .
is the
only answer.

3(i)
Write ( ) ( ) f ln 2 1 3 x x = + + .
Then,
( ) f 3
2 1 e
x
x

+ =
( )
( )
f 3
1
e 1
2
x
x

= .
Thus, ( )
( )
1 3
1
f e 1
2
x
x

= .
The domain of
1
f

is ( ) ,

or we can write as .
The range of
1
f

is
1
,
2
| |

|
\ .
.


Remark 1: Domain of
1
f

equals
to the range of f and range of
1
f

equals to the domain of f .

Remark 2: Generally, to find the
range of a function, we will need
a sketch of the graph.

HWA CHONG INSTITUTION
MATHEMATICS DEPARTMENT PAGE 3

3(ii)









The coordinates where the curves intersect the axes are:
( ) 0, 3 A= ,
( )
3
1
e 1 , 0
2
B

| |
=
|
\ .
,
( )
3
1
0, e 1
2
C

| |
=
|
\ .
and
( ) 3, 0 D = .













3(iii) When the two curves intersect, they also intersect at the
line y x = . That is, ( ) ( )
1
f f x x

= is equivalent to
( ) f x x = . Thus we have ( ) ln 2 1 3 x x + + =
( ) ln 2 1 3 x x + = .

Using the graphic calculator, the values of x are
0.4847 and 5.482, correct to 4 s.f.
Note: TI84 Plus screenshots to
get these values (based on the
intersections of the two graphs
3 y x = and ( ) ln 2 1 y x = + ) are
shown:


4(a)(i) Use integration by parts twice:
2 2 2 2 2
0 0
0
1 1
e d e 2 e d
2 2
n
n n
x x x
x x x x x

(
| | | |
=
| | (

\ . \ .

} }

2
2 2
0
e e d
2
n
n x
n
x x

= +
}

2
2 2 2
0
0
1 1
e e e d
2 2 2
n
n
n x x
n
x x

( | |
= +
| (

\ .

}

2
2 2 2
0
1
e e e
2 2 4
n
n n x
n n

(
=
(


2
2 2 2
1 1
e e e
2 2 4 4
n n n
n n

= +
( )
2
2
e 1
2 2 1
4 4
n
n n

= + + + .
Note: Usually when the integrand
consists of two different families
of terms, we should be prepared
to apply integration by parts. This
is only a guideline, as there are
exceptions such as
ln
d
x
x
x
}

where it is not necessary to
employ the technique of
integration by parts. Another
familiar exception is
3
sec d x x
}

where the usual method is to use
integration by parts to solve it.
x
y








HWA CHONG INSTITUTION
MATHEMATICS DEPARTMENT PAGE 4

4(a)(ii)
( )
2
2 2 2
0
e 1 1
e d lim 2 2 1
4 4 4
n
x
n
x x n n

| |

= + + + =
|
\ .
}
.

Note that
( )
2
2
e
2 2 1 0
4
n
n n

+ + as n .
Note 1: Integrals which have
limits ending with infinity such as
2 2
0
e d
x
x x

}
are called improper
integrals.

Note 2: A way to understand why
( )
2
2
e
2 2 1 0
4
n
n n

+ + as n

is by manipulating the expression
( )
2 2
2
2
e 2 2 1
2 2 1
4 4e
n
n
n n
n n

+ +
+ + =

and think of
( )
( ) ( )
2 3
2
2 2
e 1 2 ...
2! 3!
n
n n
n = + + + +

(according to MF15) which can
be considered an infinite sum
consisting of terms in n of much
higher degrees.
4(b)
From the diagram, the shaded region will to undergo a
rotation of one full round about the -axis.



Volume of solid obtained is given by .
1
2
0
d V y x t =
}
( )
2 1
2
2
0
16
d
1
x
x
x
t =
+
}
.
Using the substitution tan x u = :

2
d
sec
d
x
u
u
= .
When 0 0 x u = = and when 1
4
x
t
u = = .
Thus,
( )
2
4 2
2
2
0
tan
16 sec d
tan 1
V
t
u
t u u
u
=
+
}


x
y
1 0

HWA CHONG INSTITUTION
MATHEMATICS DEPARTMENT PAGE 5

2
4
2
0
tan
16 d
tan 1
t
u
t u
u
=
+
}
(since
2 2
tan 1 sec u u + = )
2
4
2
0
tan
16 d
sec
t
u
t u
u
=
}
.
2
4 4 2
2 2
0 0
sin 1
16 d 16 sin d
cos sec
t t
u
t u t u u
u u
= =
} }
. (Shown!)

Thus,
4 2
0
16 sin d V
t
t u u =
}

4
0
1 cos 2
16 d
2
t
u
t u

=
}
(by cosine double angle formula)
4
0
1
8 sin 2
2
t
t u u
(
=
(

( )
1
8 2 2
4 2
t
t t t
| |
= =
|
\ .
.


2011 GCE A Level H2 Maths Solution Paper 2
SECTION B (STATISTICS)

5
From ( ) P 40.0 0.05 X < = , we standardise X :
40.0
P 0.05
X
o o
| |
< =
|
\ .

40.0
P 0.05 Z

o
| |
< =
|
\ .

40.0
1.64485

= ----- (1)
Similarly, for ( ) P 70.0 0.975 X < = , we have
70.0
P 0.975 Z

o
| |
< =
|
\ .

70.0
1.95996

= ----- (2)
Solving (1) and (2): 53.7 = , 8.32 o = , correct to 3 s.f.

Note 1: To use the InvNorm
function in the graphic calculator,
we must ensure the probability is
of the form ( ) P ... ... X < = or
( ) P ... ... X s = , with the
inequality sign being less than
or less than or equal.

Note 2: The method of
standardisation is very efficient in
solving normal distribution
problems involving unknown
mean and/or unknown variance
(or standard deviation).
6(i)
Quota sampling might be carried out by
First divide the range of ages into several groups (12
16, 17 21, 22 26 etc).
Next, assign a quota (required number of residents to
be interviewed) for each age group.
Interviewer will then pick residents at his/her own
discretion for interviews, until all the quotas are
fulfilled.

HWA CHONG INSTITUTION
MATHEMATICS DEPARTMENT PAGE 6

6(ii)
One disadvantage is that the interviewer may likely have
collected a biased sample of residents due to his/her non-
random selection process of residents (selection is based
solely on his/her judgement).
Remark: Despite the obvious
disadvantage of quota sampling,
it is still used widely nowadays
mainly due to its big advantage of
not needing to have a sampling
frame to carry out the sampling
process.
6(iii)
Stratified sampling.

This method is not realistic in this context because it will
be extremely difficult to gain access to information about
our sampling frame. For example, it will be difficult to
get a whole list of residents with information such as
their ages from a government or public agency.




Note: Other answers are possible
such as randomly selected
residents may not be willing to
participate in the interview.

7(i)
Assumption 1: All the friends should not have any
knowledge of this experiment to be carried out, before or
during the experiment, to ensure that each trial of
contacting a friend is independent of one another.

Assumption 2: The experiment should be conducted in a
realistic span of time in the evening to ensure that time
does not become a factor to affect the constant success
probability of contacting a friend.

7(ii)
No matter how quickly the experiment can be carried
out, all the calls are bound to be done at different times in
the evening, and in turn will make it difficult for the
probability of success in contacting a friend to stay
constant.

7(iii)
For 8 n = , ( ) ~ B 8, 0.7 R and ( ) P 6 R >
( ) 1 P 5 R = s
0.552 = correct to 3 s.f.

TI84 Plus screenshots:



HWA CHONG INSTITUTION
MATHEMATICS DEPARTMENT PAGE 7

7(iv)
For 40 n = , ( ) ~ B 40, 0.7 R .

Since 40 n = is considered large, 28 5 np = > and
( ) 1 12 5 n p = > , thus, ( ) ~ N 28, 8.4 R approximately.
The parameters in this new distribution are the mean (28)
and variance (8.4).

Therefore ( ) P 25 R < ( )
.
P 24.5
c c
R <
0.114 = correct to 3 s.f.

Note: c.c stands for continuity correction.
TI84 Plus screenshots:


8(i)











Note: The points should follow a
quadratic curve of the form (in
the first quadrant):


8(ii) From the graphic calculator, the value of the product
moment correlation coefficient, r is 0.992 correct to 3
s.f.

If we observe the differences of consecutive y values
such as 18.8 16.9 1.9 = etc:

D 1.9 2.4 2.8 3.1 3.7 4.1

We would notice that differences follow an increasing
trend, which would not be the case if the data have
followed a linear model (the differences would be more
or less kept constant).
TI84 Plus screenshots to
calculating the value of r :


Note:
Below is a screenshot from TI84
Plus showing the seven observed
points and the best fit line:
y
x
HWA CHONG INSTITUTION
MATHEMATICS DEPARTMENT PAGE 8


Though the linear model is not
the best, it is not obvious from the
diagram to conclude that the
points necessary follow a non-
linear model.
8(iii)
Let
1
r and
2
r denote the product moment correlation
coefficients based on the linear model and the quadratic
model respectively. If
2 1
r r > , then we would deem the
quadratic model to be the better model. If
2 1
r r < , then
the linear model would be better.
Note: We can of course find the
value of r for the quadratic
model and do a comparison with
that obtained for the linear model.
However, since the question
asked us to explain how to use,
it will be more ideal to just
propose a general way to help
decide which model is better.
8(iv) From the graphic calculator, the equation of the least-
squares regression line of y on
2
x is
2
0.85621 22.23049 y x = + .

When 3.2 x = , the estimate value of y is
( )
2
0.85621 3.2 22.23049 13.5 y = + = correct to 3 s.f.
Some TI84 Plus screenshots:







HWA CHONG INSTITUTION
MATHEMATICS DEPARTMENT PAGE 9

9 A probability tree is drawn for the situation:


Remark: A probability tree is for
the purpose of clearer illustration.
If the question did not ask, it is
not necessary to draw a
probability tree and use it to solve
probability problems.
9(i)(a)
( ) P a lens is faulty 0.6 0.05 0.4 0.07 = +
0.058 = .

9(i)(b)
( ) P a lens is made by given that the lens is faulty A
( ) P lens is made by | lens is faulty A =
( )
( )
P lens is made by lens is faulty
P lens is faulty
A
=
and

0.6 0.05
0.058

=
0.517 = correct to 3 s.f.

9(ii)(a)
( ) P exactly one of the lenses is faulty
( )
( )
P 1st lens faulty and 2nd lens not faulty
P 1st lens not faulty and 2nd lens faulty
=
+

( ) ( ) 0.058 1 0.058 1 0.058 0.058 = +
0.109272 =
Note 1: We can also use the
complementary method to obtain
the probability:
( ) ( ) 1 P both faulty P both not faulty

Note 2: The answer in this case is
exact, thus it is not necessary to
round it off to 3 significant
figures.
9(ii)(b)
( ) P both were made by , given that exactly one is faulty A
( ) P both were made by | exactly one is faulty A =
( )
( )
P both were made by exactly one is faulty
P exactly one is faulty
A
=
and

( )( ) 0.6 0.05 0.6 0.95 2
0.109272

=
0.313 = correct to 3 s.f.

10(i)
Let denote the mean time taken to install a
component.

Null hypothesis
0
H is 38.0 = .
Alternative hypothesis
1
H is 38.0 < .

0.4
A
B
0.05
0.95
0.07
0.93
Faulty
Faulty
0.6
Not faulty
Not faulty
HWA CHONG INSTITUTION
MATHEMATICS DEPARTMENT PAGE 10

10(ii)
To reject null hypothesis, -value level of significance p s .
That is
( )
P 0.05 T t s s

38.0
P 0.05
5.0
50
t
Z
| |
|

s s |
|
|
\ .

( ) 38 50
1.64485
5
t
s
36.837 t s .
Since t denotes the sample mean time, therefore the set
of values required is { } : 0 36.8 t t e < < .

10(iii) If the null hypothesis is not rejected,
-value level of significance p > .
That is,
( )
P 0.05 T t s >
37.1 38.0
P 0.05
5.0
Z
n
| |
|

s > |
|
|
\ .

0.9
1.64485
5
n
>
83.5 n <
Since n is a positive integer, the set of values required
can be written as { } :1 83 n n e s s .

11(i)
( )
18 12
4 6
P 4 0.0941
30
10
R
| || |
| |
\ .\ .
= = =
| |
|
\ .
correct to 3 s.f.

11(ii)
Starting with ( ) ( ) P P 1 R r R r = > = + , we have
18 12 18 12
10 1 9
30 30
10 10
r r r r
| || | | || |
| | | |
+
\ .\ . \ .\ .
>
| | | |
| |
\ . \ .


18 12 18 12
10 1 9 r r r r
| || | | || |
>
| | | |
+
\ .\ . \ .\ .

( ) ( ) ( ) ( ) ( ) ( ) ( )
18! 12! 18! 12!
! 18 ! 10 ! 2 ! 1 ! 17 ! 9 ! 3 ! r r r r r r r r
>
+ + +
( ) ( )
( )
( ) ( )
( ) ( )
1 ! 17 ! 10 ! 2 !
! 18 ! 9 ! 3 !
r r r r
r r r r
+ +
>
+

( ) ( ) ( ) ( ) ( ) ( ) ( ) 1 ! 17 ! 9 ! 3 ! ! 18 ! 10 ! 2 ! r r r r r r r r + + > +
(Shown!)
Note 1:
( )
!
! !
n
n
r r n r
| |
=
|

\ .


Note 2: Most probable number
here refers to the mode of R .

Note 3: We can use TI84 Plus to
verify the correctness of r :

HWA CHONG INSTITUTION
MATHEMATICS DEPARTMENT PAGE 11

1 10
18 3
r r
r r
+
>
+

( )( ) ( )( ) 1 3 18 10 r r r r + + >
5.53 r >
Thus, the required value of r is 6 .

12(i) Let X denote the number of people joining the queue in
a period of 4 minutes. Then ( ) ~ Po 4.8 X .
Required probability ( ) ( ) P 8 1 P 7 X X > = s
0.113 = correct to 3 s.f.

12(ii) Let Y denote the number of people joining the queue in
a period of t seconds.
Then
1.2
~ Po
60
Y t
| |
|
\ .
( ) ~ Po 0.02 Y t
Given ( ) P 1 0.7 Y s =
( ) ( ) P 0 P 1 0.7 Y Y = + = = .
Thus
( ) ( )
0 1
0.02 0.02
0.02 0.02
e e 0.7
0! 1!
t t
t t

+ = .
0.02 0.02
e 0.02 e 0.7
t t
t

+ = .

From the graphic calculator, 54.87 t = .
Thus 55 t = seconds (correct to the nearest whole
number).


TI84 Plus screenshots that display
the solving of the equation:


12(iii)
Let J denote the number of people joining the queue in
a period of 15 minutes and let L denote the number of
people joining the queue in a period of 15 minutes.
Then ( ) ~ Po 15 1.2 J and ( ) ~ Po 15 1.8 L
( ) ~ Po 18 J and ( ) ~ Po 27 L .

Since in both cases, the mean number is larger than 10.
Thus, ( ) ~ N 18, 18 J and ( ) ~ N 27, 27 L approximately.
This means ( ) ~ N 18 27, 18 27 J L + approximately.
( ) ~ N 9, 45 J L
By graphic calculator, the required probability is
( ) P 11 J L > 0.617 = correct to 3 s.f.
Remark: In the process of
computing ( ) P 11 J L > , we
did not do a continuity correction
even though J L is a discrete
random variable (it follows a
Skellam distribution).
12(iv) If the time period is in terms of several hours, the rate of
people joining the check-in queue would probably not be
constant throughout, which violates one of the conditions
for a Poisson model to be valid.
Remark: The rate of people
joining the queue would certainly
be higher at a certain period of
time just before the departure of a
flight.
HWA CHONG INSTITUTION
MATHEMATICS DEPARTMENT PAGE 12