INTRODUCTION TO BUSINESS STATISTICS MARKING SCHEME

SECTION A

Q1. b Q11. a

Q2. d. Q12. b

Q3. b Q13. c

Q4. a Q14. c

Q5. c Q15. c

Q6. c Q16. a

Q7. a Q17. b

Q8. B Q18 d

Q9. a Q19. a

Q10. b Q20. b

SECTION B

Q1. (a) (i) Payback is the time in which the initial cash outflow of an
investment is expected to be recovered from the cash inflows
generated by the investment. A1

(ii) Advantage: payback period is simple to calculate. A1

Disadvantage: payback period does not take into account the

time value of money. A1

(iii) Cash flows in K millions

Year Cash flow Cumulative Cash Flow

0 (50) (50)
1 10 (40)

1
 2 13 (27)
3 16 (11)
4 19 8
5 22 30

M1 M1

So

11
=3  3.579 (or approx 4 yrs), M1, A1
19

(b) (i) The Central Limit Theorem let us use the normal distribution for
large samples. A1
This enables us calculate confidence intervals and use the
normal distribution as a test statistic in significance tests. A1

(ii) n = 50, x  32 and   6 .

Now the 96% confidence interval is:

x  2.05      x  2.05   , M1 (for finding 2.05)

n n

32  2.05  6    32  2.05  6 , M1
50 50
30.26    33.74 , M1, A1
Width of the interval: 33.74 – 30.26 =3.48, A1

(c) P(customer defaults) is p  210  0.2) , M1

(i) P(exactly 2 of 8 customers are likely to default)
= 8 C 2  0.2 2 1  0.2 = 8C2  0.2 2  0.86  0.294 , M1, A1
8 2

(ii) P(at least 2 of 8 customers are likely to default)

= 1 – P(none defaults or 1 defaults), M1


= 1  8C0  0.2 0 1  0.2
80
 8C1  0.21 1  0.2
81
, M1
= 0.496, A1

2
 (TOTAL: 20 Marks)

Q2. (a) (i) Properties of the normal probability distribution:

(I) The normal curve is bell-shaped. A1

(II) The mean, median and mode are equal. A1

(Also total area under the curve is 1; the normal curve approaches
but never touches the horizontal axis as it extends farther and
farther away from the mean)

(ii) Let O be the amount of overdraft. Then O ~ N (36000,10000 2 ) .

 O   40000   
(I) Now P(O  40000)  P   , M1
   

 40000  36000 
= P z   , z ~ N (0,1) , M1
 10000 

= Pz  0.4 = 0.5  0.1554  0.3446 , M1

Hence the number of customers with overdrafts of over K40,000 is:

0.3446  5000  1723 , M1, A1

 O   41000   
(II) P(O  41000)  P   , M1
   
 41000  36000 
= P z   , z ~ N (0,1) , M1
 10000 

= Pz  0.5 = 0.5  0.1915  0.6915 , A1

(III)
 30000  36000 40000  36000 
P(30000  O  40000)  P z ,
 10000 10000 
z ~ N (0,1) , M1

3
 = P 0.6  z  0.4 , M1

= 0.2257  0.1554  0.3811 , M1

Hence the number of customers with overdrafts between

K30,000 and K41000 is:

0.3811  5000  1,906 , M1, A1

(b)
Project A Project B
Probability Profit (K million) p x Probability Profit (K million) p x
(p) (p)

0.4 35 14 0.2 20 4
0.5 60 30 0.3 25 7.5
0.3 40 12 0.3 40 12
0.1 25 2.5 0.1 80 8
0 0.1 120 12
Total 58.5 43.5
M1, M1 M1, M1

The bank should opt for Project A on account of a higher expected value, A1

(TOTAL: 20 Marks)

Q3. (a) (i) An objective function is a linear combination of variables in a linear

problem which must be maximized or minimized. A2

(ii) Steps:

(I) Convert inequalities into equations

(II) Sketch straight line graphs

(III) Obtain a feasible region

(IV) Obtain an optimal solution by checking the value of the

objective function on the vertices of the feasible region. A4

(iii) Let l and h be number of barrels of light and heavy crude to

be bought respectively.

Oil grade Petrol Heating oil Jet fuel Cost/brl

4
 Light crude 0.3 0.2 0.3 K31500
Heavy crude 0.3 0.4 0.2 K27000
Min required 0.9 0.8 0.5

Then:

Minimise Z  31500l  27000h , M1

subject to 0.3l  0.3h  0.9

0.2l  0.4h  0.8 , M1
0.3l  0.2h  0.5 ,
l, h  0

Sketching the feasible region:

0.3l  0.3h  0.9 or l  h  3 : when l  0, h  3 and when h  0, l  3

0.2l  0.4h  0.8 or l  2h  4 : when l  0, h  2 and when h  0, l  4

0.3l  0.2h  0.5 or 3l  2h  5 : when l  1, h  1 and when

l  0 , h  2.5 , M1

M2

Checking vertices: l  0 , h  3 . Then Z  31500  0  27000  3  K81000

l  2 , h  1 . Then Z  31500  2  27000 1  K 90000

l  4 , h  0 Then Z  31500  4  27000  0  K126000

M2

Hence buy no barrel of light crude and buy 3 million barrels of heavy
crude oil to minimize cost. A1

5
 (b) Let L, M and H denote low risk, medium risk and high risk respectively.
Let F and F’ denote a policy holder files a claim and does not file a claim
respectively, M1
F
0.01
0.99 F'
L
0.6 F
0.1
0.3 M
0.1 0.9
F'
H 0.5
F

0.5 F'

M2
P H  F 
Need to find P(H|F) = , M1
P F 
0.1  0.5
= = 0.58, M1, A1
0.1  0.5  0.3  0.1  0.6  0.01
(TOTAL: 20 Marks)

Q4. (a) Situations: (i) Events occur in a space of time, A1

(ii) Events occur in a space of area or region, A1

(b) Let X be the number of customers.

(i) Then X ~ Poisson (  5 3 ) , M1

Hence P(at least two queries) = 1 – P(at most 1 query)

= 1 – P(X = 0 or X = 1) , M1

  
 e  53  5
3
0
e
5
3
 3  
 5
1

= 1   
 0 ! 1!  , M1
 
 1  0.189  0.504  1  0.693  0.307 , M1,A1
(ii) 30 queries every 45 minutes means
X ~ Poisson (  30 45) , M1

6
 Then P( X  3) =
e
30
45
 30 45

3

 0.025 , M1, A1
3!

(c) Hypotheses: H 0 : There is no association between account choice and

education level
H 1 : There is an association between account choice and
education level , M1

O  E 2
Test statistic:  
2
 E
,

Calculation of expected frequencies:

Education Account Type Total

Level Savings Current Investment Fixed
Primary 100 (123.7) 200 (185.6) 50 (39.8) 380 30 (30.9)
Secondary 300 (276.7) 400 (415.1) 80 (89) 850 70 (69.2)
Graduate 20 (19.5) 30 (29.3) 5 (6.5) 60 5 (4.9)
Total 420 630 135 1290 105
M2
Since the expected frequency in the last cell is 4.9, which is less than 5, a direct
Chi-squared analysis may be inaccurate, so we combine some categories and
calculate fresh expected frequencies for the combined categories. We combine
the first and the last rows. M1

Education Account Type Total

Level Savings Current Investment Fixed
Primary/
graduate 120 (143.3) 230 (214.9) 55 (46.0) 35 (35.5) 440

Secondary 300 (276.7) 400 (415.1) 80 (89.0) 70 (69.2) 850

Total 420 630 135 105 1290

M2
O  E  2
120  143.3 2
230  214.9 2
70  69.2 2

Then  
2
 E
=
143.3

214.9
 ..... 
69.2
= 10.059, M2

Decision Rule: Reject H 0 if    3  7.815 , M1

2 2

Conclusion: Since  2  10.059   32  7.815 , we conclude that there an

association between account choice and education level, A1
(TOTAL: 20 Marks)

7
 12
Q5. (a) Now FV  K 75,000  2  K150,000 , PV  K 75,000 and r   0.12 , M1
100
So 150, ,000  75,0001  0.12 , M1
n

2  1.12 
n

Taking logs: log 2  log 1.12  , M1

n

n log 1.12  log 2

log 2
n  6.116  6 yrs , M1, A1
log 1.12

(b) (i) The Consumer Price Index measures the annual cost of living
of a typical urban family for a particular country. A1

(ii) Let b and c be the number of kgs of beef and chicken

consumed by a typical urban family.
10
Now 2012 has  1.25 more chicken than beef basket
8
i.e. c  1.25b , M1

Now a typical cost of a basket of goods in 2010 is

500b  400c  5000 , M1
Since c  1.25b then 500b  400 1.25b  5000 , M1

500b  500b  5000 and so b  5 and c  1.25  5  6.25 , M1

These are the amounts to be included in the CPI basket for
2012 . Then the 2012 market value for 5 kgs of chicken and
6.25 kgs of beef is:
800  5  500  6.25  7125 , M1

7125
Hence the CPI is:  100  142.5 , M1, A1
5000

Interpretation: This implies that the cost of living for a typical

Malawian urban family has increased by 42.5% since the base
year. A1

(c) Hypotheses: H 0 :   35
H1 :   35 , M1

8
 x
Test statistic: Since n = 64 is large, then z  , M1

n

37  35
=  2.67 , M1
6
64

Decision Rule: At 5%, reject H 0 when z  1.64 , M1

Conclusion: Since z  2.67  1.64 , H 0 is rejected and conclude that the

mean number of cheques referred to drawer is likely more than 35. M1, A1
TOTAL: 20 Marks)