FINAL SCHEME BSM2922 SEM II 2008/2009

PART A (COMPULSORY)
Q1

(a)

The number of pounds od steam usage

600

500

400

USAGE

300

200

100
20

30

40

50

60

TEMP

F1 (for all x-axes) and F1 (for all y-axes)

Q1

(b)

Sample correlation coefficient : r =

10953.3433
= 0.9999 A1
(1182.8333)(101436.3567 )

Interpret :
Strong positive linear relationship between temperature and steam used per month. A1
Q1

(c)

x2

y2

xy

21

185.79

441

34517.9241

3901.59

24

214.47

576

45997.3809

5147.28

32

288.03

1024

82961.2809

9216.96

47

424.84

2209

180489.0256

19967.48

50

454.58

2500

206642.9764

22729

59

539.03

3481

290553.3409

31802.77

233 2106.74 10231 841161.928 92765.08

x=

y=

x = 233 = 38.8333
n

y = 2106.74 = 351.1183
n

S xx = 10231
S yy

(233)2
6

= 1182.8333 A1

2
(
2106.74 )
= 841161.928

S xy = 92765.08

= 101436.3567 A1

(233)(2106.74 ) = 10953.3433
6

A1

1 =

S xy
S xx

10953.3433
= 9.2603 A1
1182.8333

0 = (351.1183) (9.2603)(38.8333) = 8.4897 A1

^

Regression line : y = 9.2603 x 8.4897 A1

Interpret :
One unit increase in temperature (x) will increase 9.2603 in usage steam (y). A1

Q1

(d)

With x = 10 , y = 9.2603(10) 8.4897 = 84.1133 A1

Q1

(e)

Step 1 :

H 0 : 0 = 8 .3
H 1 : 0 8.3 (Claim ) B1
Step 2 :
Given : = 0.01, 2 = 0.01 2 = 0.005 , n = 6, v = 6 2 = 4
Use t-distribution, t

,v

= t 0.005, 4 = 4.604 . A1

Decision Rule :
Reject H0 if test value, falls in rejection regions, t C > 4.604 or t C < 4.604 . M1
Step 3 :
^

SSE S yy 1 S xy 101436.3567 (9.2603)(10953.3433)

MSE =
=
=
= 1.277935 A1
n2
n2
62

tT =

0 0*
_2

1 x
MSE +

n S xx

( 8.4897) ( 8.3)
1 (38.8333) 2

1.277935 +
6 1182.8333

M1

= 0.13976 A1

Step 4 :
The test value, t T = 0.13976 is less than the critical value, t C = 4.604 which is not falls in
the rejection region. We do not reject H 0 . A1
Step 5 :
We have not enough evidence to support the claim that the intercept is not equal to -8.3. A1

Q2

(a)

Type I error, is the probability of reject H0 when it is true. A1

Type II error, is the probability of do not reject H0 when it is false. A1

Q2

(b)

Step 1 :
H 0 : = 3500

H 1 : > 3500 (Claim) B1

Step 2 :

Give n = 100 , x = 3650 , = 470 , = 5% .

Condition I : known, n 30 . We use standard normal distribution. M1

The right tail test. With = 5% = 0.05 , then Z = Z 0.05 = 1.645 . A1

Decision rule :
Reject H 0 if test value falls in the rejection region, which is Z test > 1.645 . M1
Step 3 :
_

Z test =

3650 3500
470
100

M1

= 3.19148 A1
Step 4 :
Since the test value is 3.19148 which are in the critical region, we reject the null hypothesis.

A1.
Step 5 :
There is enough evidence to say that the maintenance cost will be more than RM3500 per
month. A1

Q2

(c)

Given :

Outlet A

Outlet B

Sample size

14

24

Sample mean

85

74

Sample standard deviation

12

Step 1 :
H 0 : 1 = 2

H 1 : 1 2 (Claim) B1

Step 2 :

Condition F :

12 , 22 unknown, but not equal and n1 , n2 < 30 . We use t-distribution. M1

12 2 9 2

+
14 24

v=

12
9

14 + 24
14 1
24 1
2

186.6151148
8.138147567 + 0.495244565

M1

= 21.61550315 22 A1

The two tailed test. Using = 0.1 , 2 = 0.05 , then t 0.05, 22 = 1.717 . A1

Decision rule :
Reject H 0 if test value falls in the rejection region, which is Ttest < 1.717 or Ttest > 1.717 .

M1
Step 3 :

tT =

( X 1 X 2 ) ( 1 2 )
s12 s 22
+
n1 n 2

(85 74) (0)

12 2 9 2
+
14 24

M1

= 2.97615 A1
Step 4 :
Since the test value is 2.97615, which is in the critical region, we reject the null hypothesis.

A1

Step 5 :
We can conclude that, the mean of delivery time between outlet 1 and outlet 2 is not equal.

A1
PART B (CHOICE)
Q3 (a)(i)

P( x ) = 1
k+

1 1
+ = 1 M1
2 3
k+

5
=1
6

k = 1

5
6

k=

Q3 (a)(ii)

1
6

A1

F ( 0) = P ( 0) = 0

F ( 3) = P(0) + P(1) = 0 + 1 6 = 1 6

M1

F (6) = P(0) + P(1) + P( 2) = 0 + 1 6 + 1 2 = 2 3

F (9) = P(0) + P (1) + P( 2) + P(3) = 0 + 1 6 + 1 2 + 1 2 = 1
0
1 / 6

F ( x) =
2 / 3
1

Q3 (a)(iii)

M1

x < 3

, 3 x < 6
,

6 x<9

x9

A1

1 1 1
E ( X ) = 3 + 6 + 9 M1
6 2 3
= 5.5 A1
1
1
21
E ( X 2 ) = ( 3) + 6 2 + 9 2 M1
6
2
3
= 46.5 A1

E [( X + 2) 2 ] = E [( X 2 + 4 X + 4)] = E ( X 2 ) + 4 E ( X ) + E ( 4) = 46.5 + 4(5.5) + 4 M1

= 72.5 A1

Q3 (a)(iv)

Var ( X ) = E ( X 2 ) [E ( X )] = 46.5 (5.5) 2 = 16.25 A1

2

Var (4 X + 1) = 4 2 Var ( X ) + Var (1) M1

= 16(16.25) + 0 M1
= 260 A1
+

Q3 (b)(i)

f ( x ) dx = 1

1 x 5
(0) dx + 0 5 e dx = 1 M1

1 x 5
e
dx = 1
5 0

x 5

1 e

= 1 M1
5 1

5 0

[ e ] = 1
] [ e ] = 1
x 5
0

[ e

( ) 5

0 5

(0) ( 1) = 1
1 = 1 (Prove) A1

Q3 (b)(ii)

P ( 1 X < 2 ) =
0

1 x / 5
e
dx
5

1 x 5
e
dx
5 0
2

1 e x 5
=

5 1
5 0

[
]
] [ e ]
= [ e
= e x 5

2
0

( 2) 5

05

M1

= ( 0.67032) ( 1)
= 0.32968 A1

Q4

(a)(i)

Given that n = 10 and p = 0.08 , then X ~ B (10, 0.08) B1

P ( X = 10) =

10

C10 0.0810 0.921010 M1

= 1.07374 10 11 A1

Q4

(a)(ii)

= np
= 10 0.08 M1
= 0.8 A1

Q4

(b)(i)

Given that, X ~ P0 ( 2.2)

Then, 2.2 1 m 2 and x 10 m 2

x 2 .2
=
, then x = 2.2 10 = 22
10
1

X ~ P0 ( 22)
P ( X < 20) = 1 P ( X 20) M1
= 1 0.6940
= 0.3060 A1

M1

Q4

(b)(ii)

Then, 22 10 m 2 and x 100 m 2

x
22
=
, then x = 22 10 = 220
100 10

M1

X ~ P0 ( 220)

= 220 A1

2 = 220 A1
Q4

(c)(i)

Given that : X ~ N (0, 1) B1

1.32 0

M1 P ( X 1.32 ) = P Z <
M1
1

= P ( Z < 1.32)
= 1 P ( Z > 1.32) M1
= 1 0.0934
= 0.9066 A1

Q4

(c)(ii)

1.02 0
0 .3 0
M1 P ( 0.3 < X < 1.02 ) = P
<Z<
M1
1
1

= P( 0.3 < Z < 1.02 )

= 1 P ( Z > 0.3) P ( Z > 1.02) M1
= 1 0.3821 0.1539
= 0.4640 A1

Q5

(a)(i)

2510 2500
2499 2500
P( 2499 < X < 2510) = P
<Z<
M1
2
2
50
50

5
5

= P ( 0.04 < Z < 0.45)

= 1 P ( Z > 0.04) P ( Z > 0.45) M1
= 1 0.4840 0.3264 M1
= 0.1896 A1

Q5

(a)(ii)

2550 2500

P ( X < 2550) = P Z <

M1
500

= P ( Z < 2.24)
= 1 P ( Z > 2.24) M1
= 1 0.01255

= 0.9875 A1

Q5 (a)(iii)

2450 2500

P( X 2450) = P Z >
M1
2
50

10

= P ( Z > 3.16)
= 1 P ( Z > 3.16) M1
= 1 0.00069
= 0.9993 A1

Q5

(b)(i)

P (X H

Low Concentration

High Concentration

Sample mean

55

60

Sample standard deviation

Sample size

16

16

58 60

> 58) = P Z >

M1
2
4

16

= P (Z > 2 )
= 1 P (Z > 2 ) M1
= 1 0.02275
= 0.97725 A1

Q5

(b)(ii)

54 55

P (X L < 54 ) = P Z <
M1
2
4

16

= P (Z < 1)
= P (Z > 1) M1
= 0.1587 A1

Q5 (b)(iii) Step 1 :

H L = 60 55 = 5
Step 2 :

42 42
=
+
=2
16 16
2
H

2
L

Step 3 :

X H X L ~ N (5, 2 )
Step 4 :
0 5

P ( X H X L > 0) = P Z >
M1
2

= P ( Z > 3.54)
= 1 P ( Z > 3.54) M1
= 1 0.0023 M1
= 0.9977 A1

Q6

(a)(i)

Given that : E = 20 and = 100 .

Level of confidence : (1 )100% = 99%

= 1 0.99 = 0.01

2 = 0.01 2 = 0.005
From standard normal table : Z 2 = Z 0.005 = 2.58
Z 2
Sample size, n =

( 2.58) (100)
n=

20

M1

n = 166.41
n 167 A1
So we will need to sample at least 167 (rounded up) randomly. With this sample we will be
99% confident that the sample mean will be within 20 of the true population.

Q6

(a)(ii)

Step 1 :
Given, n = 27 and = 100 (sample)

Condition B : known, n < 30 , we use normal distribution. M1

Step 2 :
Level of significance : (1 )100% = 98%

= 1 0.98 = 0.02

2 = 0.02 2 = 0.01

From standard normal table : Z 2 = Z 0.01 = 2.33 M1

Step 3 :

The point estimate, x = 290

Step 4 :
_
_


x z 2
< < x + z 2

n
n

100
100
290 2.33
< < 290 + 2.33
M1
27
27
245.1591 < < 334.8408 A1
Step 5 :
With 98% confidence level, we can conclude that based on 27 televisions, the true mean
value of all television is between 245.1591 and 334.8408. A1

Q6

(b)(i)

Step 1 :
Academic Staff

Administration Staff

n1 = 14

n 2 = 17

x 1 = 9 .2

x 2 = 7 .9

s1 = 1.3

s 2 = 1.5

Case D : 12 = 22 unknown, n1 n 2 and we use t-distribution.

Step 2 :
Level of significance : (1 )100% = 95%

= 1 0.95 = 0.05

2 = 0.05 2 = 0.025
Degree of freedom, v = n1 + n2 2 = 14 + 17 2 = 29
t 2,v = t 0.025, 29 = 2.045 A1
S

2
p

(
14 1)(1.3) 2 + (17 1)(1.5) 2
=
14 + 17 2

M1

S p2 = 1.99896 A1
Step 3 :
_

Point estimate : x 1 x 2 = 9.2 7.9 = 1.3

Step 4 :
Confidence interval for difference means :
1
1
1
1
_ _
_ _
+ < 1 2 < x 1 x 2 + t 2,v S p
+
x 1 x 2 t 2,v S p

n1 n 2
n1 n 2

(9.2 7.9 ) (2.045)(

1.99896

1
1
+
< 1 2 < (9.2 7.9 ) + (2.045) 1.99896
14 17

1
1
+
14 17

M1

0.25651 < 1 2 < 2.34348 A1

Step 5 :
The difference between means time of attendance record by academic staff to administration
staff is between 0.25651 and 2.34348. A1

Q6

(b)(ii)

Step 1 :
_

x=

x = 1599 + 1699 + 1799 + 1899 + 1999 + 2099 + 2299 + 2499 + 2699

n

18591
9

= 2065.67 A1
n x 2 ( x )

s =
2

n (n 1)

9(39502809 ) (18591)
=
9(9 1)

= 137500 A1
We use chi-square distribution. M1
Step 2 :
Level of significance : (1 )100% = 98%

= 1 0.98 = 0.02

2 = 0.02 2 = 0.01
Degree of freedom, v = n 1 = 9 1 = 8

2 2,v = 2 0.01,8 = 20.09

21 2, v = 2 0.99,8 = 1.646
Step 3 :
The point estimate, s 2 = 137500

A1

Step 4 :

(n 1) s 2 < 2 < (n 1) s 2
2
2

2, v

1 2 , v

(9 1)(137500) < 2 < (9 1)(137500)

20.09

1.646

M1

54753.608 < 2 < 668286.755 A1

Step 5 :
With 98% of confidence level, we can say that the true variance of the population is between
RM54753.608 and RM668286.755. A1