QUANTITATIVE TECHNIQUES

TOPICS
1. Data collection and presentation (UNGROUPED DATA)
2. Measures of location and dispersion (GROUPED DATA)
3. Index numbers
4. Simple liner regression and co-relation
5. Probability distribution
6. Financial mathematics (Simple interest, Annuities, Compound)
7. Sampling theory
8. Confidence interval and hypothesis testing
9. Time series
10. Calculus
11. Matrix
12. Approximations and errors
13. Linear programming
14. Project management

Requirements

 Scientific calculator
 Syllabus
 Formulae list and tables

Text books

 Applied statistics by Trevor Wegner

 ZOU – Business Statistics Hand Book/Modules

MEASURES OF LOCATION AND DISPERSION

What to measure in ungrouped data?

 Mean
 Mode
 Median

1|Page
  Geometric mean
 Harmonic mean

Formulae

e.g. 10 60 65 61 56 72

x
1. Mean, x = n

10  60  65  61  56  72
= 6

= 54

2. Mode = it is the most highest frequently of applied data

e.g. 10 60 65 61 56 72

= 72

3. Median = it is a middle value of the ordered data

Steps of median:

Arrange your data in increasing or decreasing order

Determine the median position

1
n  1
Median position = 2

Hence state the median

e.g. 10 60 65 61 56 72

= 10 56 60 61 72

1
n  1
Median position = 2

2|Page
 1
6  1
= 2

= 3, 5

60  61
 Median = 2

= 60, 5

OR ALTERNATIVELY

Median = 60+0.5(61-60)

= 60+0.5(1)

= 60.5

4. GEOMETRIC MEAN (G.M) %

It is associated with percentages changes

n x1  x2  x3  ...  xn
GM =

n= number of percentages changes

e.g. over a four (4) year period, inflation rates were as follows; 2% 3.5%
4.1% and 5.2% respectively. Calculate the average inflation rate per annum.

NB; You must add all percentages with 100%

n x1  x2  x3  ...  xn
GM =

=
4
102  103.5  104.1  105.2

= 103.69

NB; then you must subtract your answer with 100%

 Average inflation rate = 3.69%

Electricity tariffs have increased by 12%, 8% and 16% per annum over the three year
period. Using a suitable method, find the annual average in the electricity.

n x1  x2  x3  ...  xn
GM =

3|Page
 =
3
112  108  116

= 111.9523607

= 11.95% (2 d.p)

5. HARMONIC MEAN (HM)

It is associated with rates of change e.g. speed, birth rate, growth rate

n
1
x1
 1
x2
 ...  1
xn
HM =

Five (5) communiter omnibuses A, B, C, D and E are travel from Marondera to Harare
at a speed of 90km/hr, 100km/hr, 110km/hr, 120km/hr and 130km/hr respectively.
Using a suitable method, calculate the average speed of the buses.

n
1
x1  1
x2  ...  1
xn
HM =

5
1
90  1
100  1
110  120
1
 130
1
=

= 108.16KM/H

6. WEIGHTED MEAN (WM)

 fx
WM, = f
Height (x) 1.5m 1.6m 1.7m 1.8m 1.9m
Frequency (f) 2 3 7 6 2

 fx
WM = f
2(1.5)  3(1.6)  7(1.7)  6(1.8)  2(1.9)
= 23 7  6 2

4|Page
 = 1.715m

A lecturer is paid $50 per hour for an 8hour session at college X; he is also paid $150
per hour for a 3hour session at college Y; and $200 per hour for a 2hour session at
college Z. Calculate the lecturer’s average hourly income from three colleges.

College X Y Z
Hourly rate $50 $150 $100
Time taken 8hours 3hours 2hours

{8(50)  3(150)  2(100)}

WM = {8  3  2}

= $96.15/HOUR (2 d.p)

EXAMPLE 2

The exam marks of a class of QT students were recorded as follows

80 40 50 80 70 50 65 75 83 92 97 50 82
93 97 99 49 78 82 73 79 77 85 79

Calculate; mean, mode, median (20 MARKS)

x
 Mean, x = n

80  40  50  ...  79
= 24

= 75.2083

 Mode = 50
1
n  1
 Median = 2

1
24  1
= 2

= 12.5

NB; Arrange your data in decreasing or increasing order

5|Page
 79  79
 Median = 2

= 79

MEASURES OF DISPERSION or SPREAD

What to study?
 Range (the difference between the maximum and minimum)
 Interquartile range (IQR)
 Mean absolute deviation
 Variance
 Standard deviation
 Coefficient of variance

Range

It is the different between the maximum and minimum. Referring from the previous
example on marks obtained by 24 QT student, calculate the range.

Range = 99 – 40

= 59

Variance

Sample variance Population variance

 x2 n x   x2  u 2
2

n 1 n

 x x   xu 2
2

n 1 n

Using data on marks obtained by QT students; calculate

Variance, Standard deviation and Coefficient of variance (20 marks)

Solutions

6|Page
  x2 n x 
2

i. Variance = n 1

142169 24 75.20833 2
241
=

= 279.04219

ii. Standard deviation, (S) =

var iance

 x2 n x 
2

n 1
=

142169 24 75.20833 2
241
=

= 16.7045

s
 100
iii. Coefficient of variance (CV) = x
16.70
75.21  100
=

= 22.2045%

GROUPED DATA
Measures of location or central tendency

 fx
i. Mean, x = f
C[ fm  fm  1]
X mod 
ii. Mode = 2 fm  fm  1  fm  1

7|Page
 Key where

Xmod = lower limit of the modal interval

C = class width of modal interval

Fm = represents the frequency of modal interval

Fm-1 = represents the frequency just before the modal interval

Fm+1 = the frequency just after the modal interval

iii. Median, (Q2)

C[ n2  f  ]
Xmed 
fmed
Median, (Q2) =

Key

Xmed = lower limit of the median class interval

n
2
= median position

F(<) = represent the cumulative frequency just before the median interval

f med = frequency of the median interval

C = class width of the median class interval

EXAMPLE 1

Given below is the data concerning age of the borrowers in purchasing a house by means of
mortgage.

Age (years) % of borrowers

15 – 25 21

8|Page
25 – 35 45
35 – 45 21
45 – 55 10
55 – 65 3
Calculate:

Mean, mode, median (12 marks)

Solutions

Age (years) Mid-point (x) % of borrowers fx cumulative

(f)
15 – 25 20 21 420 21
25 – 35 30 45 1350 66
35 – 45 40 21 840 87
45 – 55 50 10 500 97
55 – 65 60 3 180 100

100 (∑f) 3290 (∑fx)

 fx
i. Mean x = f

3290
= 100

= 32.9

C[ fm  fm  1]
X mod 
ii. Mode = 2 fm  fm  1  fm  1

25  21045( 45212121)
=

9|Page
 = 30

C[ n2  f  ]
Xmed 
fmed
iii. Median, (Q2) =

25  10 (504521)
=

= 31.4444

C[ n4  f  ]
Xq1 
fq1
iv. Lower quartile, (Q1) =

25  10 ( 254521)
=

= 25.8889

C[ 34n  f  ]
Xq3 
fq3
v. Upper quartile, (Q3) =

35  10 ( 752166 )
=

= 39.2857

Q3  Q1
vi. Quartile deviation = 2

39.2857  25.89
= 2

= 6.7

EXAMPLE 2

The following data gives the salary earned by a sample of 40 employees at a certain
manufacturing company

10 | P a g e
Salary (US$) Number of employees
200 – 400 3
400 – 600 10
600 – 800 17
800 – 1000 6
1000 – 1200 4

Calculate: mean, quartile deviation, median, 70th percentile, 6.5th decile (20)

Solution

EXAMPLE 3

The weights of fish in a catch are as follows

Weight (kg) Number of fish

0-<2 5
2-<4 12
4-<6 18
6-<8 13
8 - < 10 2

Calculate: coefficience of variance (CV), mode, median, coefficience of skewness

(20marks)

Solutions

Weight (kg) Mid-point (x) Frequency (f) fx fx2 Cumulative

0–2 1 5 5 5 5
2–4 3 12 36 108 17
4–6 5 18 90 450 35
6–8 7 13 91 637 48
8 – 10 9 2 18 162 50

50 (  240 (  1362 ( 
f fx fx2
) )
)

11 | P a g e
 s
 100
CV = x

 fx
i. Mean, x = f
240
= 50

= 4.8kg

 fx2  n x 
2

ii. Variance = n 1

1362  504.8
2

= 50  1

= 4.28571

iii. Standard deviation, S = var iance

=
4.28571

= 2.0702

S
 CV = x
 100

2.0702
 100
= 4.8

= 43.1292%

C[ fm  fm1 ]
Xmod 
2 fm  fm1  fm1
i. Mode =

2(18  12)
4
= 2  18  12  13

12 | P a g e
 = 5.0909

C[ n2  f  ]
Xmed 
fmed
ii. Median =

225  17 
4
= 18

= 4.8889

3mean  median 
 100
iii. Coefficient of skewness = s

34.8  4.8889 
= 2.0702
= - 0.1288

Comment: There is a weak negative association between weight and number of fish.

KURTOSIS
It is the degree of peakedness of a distribution

 Leptokurtic – it is highly peaked distribution

 Mesokurtic – it is moderately peaked distribution and it is more of a normal
distribution
 Platykurtic – it is small spread out of the distribution, it is more or less spread out of
distribution

INDEX NUMBER
It is a percentage change of a given variable (item) between the two time periods.

Simple index

There are 3 types of simple index namely:

13 | P a g e
 i. Simple Price Index or Simple Price Relative (SPI)

P1
 100
p0
SPI =

p1
= Key - is the current period price

p0
Is the base period price
ii. Simple Quantity Index or Simple Quantity Relative (SQI)

Q1
 100
Q0
SQI =

Q1
KEY is the current period year

Q0
Is the base period quantity

iii. Simple Value Index (SVI)

P1Q1
 100
P0Q0
SVI =

It has current period price and quantity over or divided by base period of price and quantity,
then multiplies it by 100.

Aggregate indexes
There are 2 types of aggregated indexes namely:

 Unwaited aggregated index

 Waited aggregated index

1. UNWAITED AGGREGATED INDEX

There are 2 types of Unwaited aggregated index namely

i. Unwaited aggregated price index (UAPI)

 P 100
1

UAPI = P 0

14 | P a g e
 ii. Unwaited aggregated quantity index (UAQI)

Q 1
 100
UAQI = Q 0

2. WAITED AGGREGATED INDEXES

There are two 2 approaches:

 Laspeyers Approaches (Base Period Weights)

 Paache Approach (Current period weight)

a) LASPEYRES APPROACH

Advocates holding or fixing weights at the base period level

i. LASPEYERES PRICE INDEX (LPI)

Quantities are taken as weights at the base period level

 PQ1 0
 100
LPI = PQ 0 0

ii. LASPEYERES QUANTITY INDEX (LQI)

Prices are taken as weights at the base period level

PQ0 1
 100
LQI = PQ0 0

b) PAACHE APPROACH

Advocates holding or fixing weight at the current period

i. PAASCHE PRICE INDEX (PPI)

Quantity is taken as weights at the current period

 PQ1 1
 100
PPI = PQ 0 1

15 | P a g e
 ii. PAASCHE QUANTITY INDEX (PQI)

Price is taken as weight at the current period

 PQ1 1
 100
PQI =  PQ1 0

FISHERS PRICE INDEX (FPI)

FPI = LPI  PPI

FISHERS QUANTITY INDEX (FQI)

LQI  PQI
FQI =

EXAMPLE 1

The table below shows the prices and quantity of 4 products between 1996 and 1998

PRODUCT PRICE $ QUANTITY (KG)

1996 1998 1996 1998
A 95 148 260 25
B 200 305 124 118
C 92 122 100 120
D 50 98 48 45

Using 1996 as base year calculate:

i. The price relative of product A

ii. Unweight indexes
iii. Weighted price index
iv. Fisher’s price index
v. Fishers quantity index (20)

16 | P a g e
SOLUTIONS

P1
 100
P0
i. SPI =

148
 100
= 95

= 155.7895

Comment: the price of product A was increased by 55.7895% between 1996 and 1998

 P 100
1

ii. UAPI = P 0

148  305  122  98

 100
= 95  200  92  50

= 154.0046

Comment: there was an average price increase of 54.0046% between 1996 and 1998

Q 1
 100
iii. UAQI = Q 0

25  118  120  45
 100
= 260  124  100  48

= 57.8947

COMMENT: There was an average quantity increase of 57.8947% between 1996 and 1998

 PQ 1 0
 100
iv. LPI = PQ 0 0

148  260  305  124  122  100  98  48

 100
= 95  260  200  124  92  100  50  48

= 152.5434

COMMENT: There was holding quantities constant at the base period level, prices increases
on an averages on an average of 52.53% between 1996 and 1998

17 | P a g e
  PQ
1 1
 100
v. PPI = PQ0 1

148  25  305  118  122  120  98  45

 100
= 95  25  200  118  92  120  50  45

= 149.5989

COMMENT: There was a holding price constant at the base period levels prices increase on
an average of 49.6% between 1996 and 1998

vi. FPI = LPI  PPI

= 154.5434  149.5989

= 151.0639

 PQ1 1
 100
vii. VAULE INDEX = PQ 0 0

148  25  305  118  122  120  98  45

 100
= 95  260  200  124  92  100  50  48

= 96.1375

FIXED BASED INDEXING METHOD

The common based period is used. It is used to monitor long term trade as references is
made to a constant base.

FIXED BASE PRICE INDEX (FBPI)

Pn
 100
P0
FBPI =

18 | P a g e
FIXED BASE QUANTITY INDEX (FBQI)

Qn
 100
Q0
FBQI =

Consider the following enrolment figure at a particular college

Year 2008 2009 2010 2011 2012

Enrolment 255 310 389 470 450

Calculate the fixed based indexes, taking 2008 as the base year

Solutions

255
 100
2008: 255 = 100

310
 100
2009: 255 = 121.568

19 | P a g e
 389
 100
2010: 255 = 152.55

470
 100
2011: 255 = 184.31

450
 100
2012: 255 = 176.47

Year 2008 2009 2010 2011 2012

FBI 100 121.57 152.55 184.31 176.47

CHANGE OF BASE YEAR

Step1

Calculate adjustment factor (AF)

100
AF = old * index * of * new * base * year

Stpe2

Multiply all the old indexes by the AF to obtain the new indexes

Consider the table below

Year 2008 2009 2010 2011 2012

FBI 100 121.57 152.55 184.31 176.47

Change base year to base 2010

100
AF = old * index * of * new * base * year

100
= 152.55

= 0.65552

2008: 0.65552  100 = 65.55

2009: 0.65552  121.57 = 79.69

20 | P a g e
2010: 0.65552  152.55 = 100

2011: 0.65552  184.31 = 101.1

2012: 0.65552  176.47 = 115.68

YEAR 2008 2009 2010 2011 2012

FBI 65.55 79.69 100 101.1 115.65

CHAIN BASE METHOD INDEX

It is associated with a shifting method i.e. the base year is always in preceding time period.

Pn
 100
Pn  1
CBI =

Referring to the data below, calculate CBI

Year 2008 2009 2010 2011 2012

Enrolment 255 310 389 470 450

21 | P a g e
SOLUTION

Pn
 100
Pn  1
CBI =

2008 : –

310
 100
2009 : 255 = 121.57

389
 100
2010 : 310 = 125.48

470
 100
2011 : 389 = 120.82

450
 100
2012 : 470 = 95.74

YEAR 2008 2009 2010 2011 2012

CBI - 121.57 125.48 120.82 95.74

WEIGTED PRICE INDEX (WPI)

 PW 100
WPI = PW 0

Consider the following table, calculate WPI

COMMODITY YEAR 2007 YEAR 2008

WEIGHTING PRICE PRICE
A 2 0.20 0.16
B 1 0.75 1.50
C 2 0.80 1.20
D 3 1.00 2.00

SOLUTION

 PW 100
WPI = PW 0

22 | P a g e
 0.16  2  1.50  1  1.20  2  2.00  3
 100
= 2  0.20  1  0.75  2  0.80  3  1.00

= …………….??????????????

SIMPLE LINEAR REGRESSION AND CORELATION

Regression analysis

It refers to establishment of a function relationship between the dependent variable and

the independent variable.

Correlation analysis

It involves the assessment of the degree or strength of association between the dependent
and independent variable.

Common terms used are:

Dependent variable; it refers to that variable whose behaviour is influenced or affected by

another variable.

Independent variable; it refers to that variable whose behaviour is not influenced by

another variable.

SCATTER DIAGRAMS

PERFECT POSITIVE LINEAR ASSOCIATION

REGRASSION EQUATION

It is used for predictions purposes

It is given that : Y = a + bx

23 | P a g e
 n xy   x y
n x2   x
2

b =

 y  b x
a = n

PEARSON’S CORELATION COEFFICIENT (r)

It is used to quantify the degree of the association between the dependent and
independent variable

It is a measure of strength of association between the dependent and independent variable

It ranges from (– 1 to + 1)

n xy   x y
{n x2   x }{n y2   y }
2 2

r =

Example

MONTH OUTPUT IN THOUSAND UNITS (x) COST IN THOUSAND DOLLARS (y)

1 20 82
2 16 70
3 24 70
4 22 85
5 18 73

Find equation to determine the expected level of cost for any given (10)

Find output when it is $22 000

SOLUTION

Y = a  bx

n xy   x y
n x2   x
2

b =

58104  100  400

5  2040  100 
2
=

24 | P a g e
 = 2.6

 y  b x
a = n

400  2.6  100

= 5

= 28

 Y = 28  2.6 x

Output when it is $22 000

Y = 28  2.6 x

= 28  2.6  22

= $85.2 thousand

n xy   x y
{n x2   x }{n y2   y }
2 2

r =

5  8104  100  400

[5  2040  100  ][5  32278  400  ]
2 2
=

= 0.9862

Comment; there is a very strong positive linear association between output and cost of
production.

RS )
SPEARMAN’S RANK CORRELATION COEFFICIENT (

6 d 2
1
(
RS )
= 
n n2  1 
Key

25 | P a g e
d = represents or is the difference per paired rankings

n = number of pairs

Example

A group of O’Level students was tested in maths and English subjects.

STUDENT A B C D E F G H
MATHS RANKING 2 7 6 1 4 3 5 8
ENGLISH RANKING 3 6 4 2 5 1 8 7

Find the spearman’s rank correlation coefficient for the above data and comment it.
(10)

SOLUTION

STUDENT A B C D E F G H total
MATHS RANKING 2 7 6 1 4 3 5 8
ENGLISH RANKING 3 6 4 2 5 1 8 7
D -1 1 2 -1 -1 2 3 1
d2 1 1 4 1 1 4 9 1 22

6 d 2
1
 Rs
= 
n n2  1 
6  22
1
= 
8 82  1 
= 0.7381

Comment; there is a strong positive association between performance in maths and English

PROBABILITY DISTRIBUTIONS
It is the chance that a particular event will be occurring.

BASIC FORMULA FOR PROBABILITY

Consider the event A from a sample space (possibility spaces)

26 | P a g e
 n( A)
P(A) = n( S )

BASIC PROPERTY OF PROBABILITY

0 is less or equal to P(A) or greater or equal to 1

If P(A)=0, then the event A is an impossible event

If P(A)=1, then the event A is a certain event

P ( A)
=1-P(A)

GENERAL ADDITIONAL OF PROBABILITY

For any events A and B, probability of (AorB) = P(A)+P(B)-P(A and B)

P(A or B) = P(A) + P(B) – P(A and B)

SPECIAL ADDITIONAL RULE (INDEPENDENT EVENTS)

Events A and B are independent event, (iff) the occurrence and non-occurrence of the event
A does not affect the occurrence of non-occurrence of B.

Hence,
P ( A B)  P ( A)  P ( B)

The number and gender of employees of NGO, INTERNATIONAL and INDIVIDUALS

companies were recorded in a tabulated below.

GENDER TYPES OF TOTALS

COMPANIES
INTERNATIONAL INDIGENOUS NGOs
MALE 200 200 100 500
FEMALE 150 300 100 550
350 500 200 1050

Find the probability that:

I. A random selected is a female

II. Is employed by an NGOs
III. Is a male employed by an international company
IV. What percentages of employees of NGOs are employed by indigenous company

Find the probability that is employed selected at random is:

i. A male or employed by an indigenous

27 | P a g e
 ii. Is a female given that she is employed by an international company
iii. Is a male given that he is employed by an NGOs

SOLUTIONS

550 11
P(female) = 1050 = 21

200 4
P(employed by an NGOs) = 1050 = 21

200 4
P(male employed  International company) = 1050 = 21

0
P ( NGOs  indigenous) = 1050 = 0%

CONDITIONAL PROBABILITY

Let A and B be any 2 events, then the conditional probability event A given that event B has
occurred, then it is the jointly probability of event A and B occurring divided by the
probability that event B has occur is equally.

P ( A/ B) 
∩
P ( A B)
P ( B)

Of 250 employees of a company, a total of 130 smoke cigarettes. There are 150 males
working for this company and 85 of them are males who smoke cigarettes.

a) Construct a contingents table

b) What is the probability that an employee chosen at random:
i. Does not smoke cigarettes given that she is a female
ii. Is male given that is smokes cigarettes?
iii. Determine whether cigarette smoking’s being male are statically independent

SOLUTIONS

Contingent table

SMOKING HABIT GENDER TOTAL

MALE FEMALE
SMOKERS 85 45 130

28 | P a g e
NON-SMOKERS 65 55 120
TOTAL 150 100 250

n(non  smo ker s )

P(non-smokers) = No  of  employees

120
= 250

12
= 25

45
P ( Female ∩smokes) = 250

9
= 50

195
∩
P (male smokes) = 250

39
= 50

OR =
∩
P ( MALE)  P ( SMOKE)  P ( MALE SMOKE)

150 130 85
 
= 250 250 250

1 253  250
85
=

39
= 50

OR =
∩
P (male smokes)

85  65  45
= 250

39
= 50

29 | P a g e
 P
∩
( Non  smo ker s female)
P ( Non  smo ker s / female) = P ( Female )

55  100 
 
= 250  250 

11
20
=

∩
P (male smokes)
P (male / smokes) P ( Smokes)
=

85 130

250 250
=

17
= 26

NB - Two events A and B are independent

P AandB P A P B

=

85
P Male & Smoker s  250
=

17
50
=

P male  P Smoker s 
OR =

30 | P a g e
 150 130

250 250
=

39
125
=

17 39

50 120 
SINCE = Are not independent

TREE DIAGRAMS

If two cards are drown at once from an ordinary well shuffled deck. What is the probability

will be:

Both will be queens

One will be a queen and the other is not?

Neither will be a queen

SOLUTIONS

4 3

52 51
P(Both will be queen) =

1
221
=

4 48 48 4
  
52 51 52 51
P(one will be a queen and the other not) =

31 | P a g e
 32
221
=

P (QQ)
P(Neither will be a queen) =

48 47

52 51
=

PROBABILITY DISTRIBUTIONS

Discrete distributions (Exact value)

a) Binomial distribution

b) Poisson distribution

c) Continuous distribution (any value)

THE BINOMIAL DISRIBUTION

Characteristics / features / properties

i. The experiment consists on “n” independent trials

ii. Each trial results in one of two possible mutually exclusive outcomes called “success”

and “failure”

iii. The probability of “success” ‘p’, remains constant from trial to trial and the

32 | P a g e
 probability of “failure” ‘q’ is such that q=1-p

iv. Probability of “success” and “failure”=1

v. P+Q=1

The binomial random variable “x” is the number of success in any trials.

Formula

If “x” is a binomial random variable, so it is associated with any independent trials

P X  r  Cr p r q nr
n

Example

In an Agricultural cooperative claims that 95% of the water melons that are shipped out are

rape and ready to eat. Find the probability that:

i. Among 10 water melons are randomly to be shipped out

ii. There is not one is rape and ready to eat

iii. At least 8 are rape and ready to eat

iv. Between 6 and 10 are rape and ready to eat

v. At most 8 are rape and ready to eat

vi. At least one is rape and ready to eat

vii. Exactly 2 are not rape and ready to eat

viii. At most one is not rape and ready to eat

SOLUTIONS

33 | P a g e
NB

Exactly =


At least / minimum


At most / maximum

Greater than / more than >

Less than <

SOLUTIONS

P=0.95 q=0.05 n=10

n
Cr p r q nr
P(x=0) =

10
C0  0.950  0.05100
=

= 0.00000000000009766

px  8
= P(8)+P(9)+P(10)

n
Cr p r q nr
=

10
C8  0.958  0.05108 10C9  0.959  0.05109 10C10  0.9510  0.051010
=

34 | P a g e
 = 0.9885 (4 d.p)

P(6<x<10) = P(7)+P(8)+P(9)

10
C7  0.957  0.05107 10C8  0.958  0.05108 10C9  0.959  0.05109
=

p ( x  8)
= P(0)+P(1)+…+P(8)

= 1 - {P(9)+P(10)}

1 [10C9  0.959  0.05109 10C10  0.9510  0.051010 ]

=

P ( x  1) P (1)  P (2)  ...  P (10)

=

= 1 – P(0)

1 [10C0  0.950  0.05100 ]

=

35 | P a g e
NB P=0.05 q=0,95 n=10

P(x = 2) = P(2)

10
C2  0.052  0.95102
=

P ( x  1)
= P(0)+P(1)

10
C0  0.050  0.95100 10C1  0.051  0.95101
=

NB

If “x” is a binomial random variable, then:

Mean = np

Variance = npq

POISSON DISTRIBUTION

36 | P a g e
Properties or characteristics

i. The occurrence of events are independent i.e. occurrence of an event in an interval

in a space / time has no effect on the probability of other event in the same / other

interval

ii. The probability of a single occurrence of the event in a given interval is proportional

to the length of the interval

iii. The mean number of events in each unit will be de-noted by “a”

iv. If “x” is a Poisson random variable, then the probability distribution formula is given

as follows

e  a .a x
x!
P(X = x) =

= x=0,1,2…

NB

Mean = a (number of occurrence)

Variance = a

Examples of Poisson situations

i. The number of calls received by a switchboard operator in an hour

ii. The number of customers received by teller in an hour

37 | P a g e
 iii. The number of death per day due to rare disease in a year

iv. The number of operating machines that breakdown in a plant during a month

Example

Suppose the number of complaints received at a company per day is a random variable

having the Poisson distribution with mean = 4

What is the probability on a given day?

Probability that is no complains will be received

Probabilities that there is no more than 1 complain will be received

There is probability of at least 3 complains will be received

What is the probability that on a given half day:

There is no complaints will be received

In between 1 and 4 complaints will be received

SOLUTIONS

Mean = 4 per day

38 | P a g e
 e 4  40
0!
P(x = 0) =

= 0.0183

P x  1
= P(0)+P(1)

e 4  40 e 4  41

0! 1!
=

= 0.0916

P x  3
= P(3)+P(4)+…

1  [ P 0  P (1)  P (2)]
=

e 4  40 e 4  41 e 4  4 2
1[   ]
0! 1! 2!
=

= 0.7619

Mean = 2 per half day

e 2  20
0!
P(x = 0) =

39 | P a g e
 = 0.1353

P(1 < x < 4) = P(2) + P(3)

e 2  2 2 e 2  23

2! 3!
=

= 0.4511

CONTINUOUS DISTRIBUTION

Normal distribution

Properties

It is bell shaped

It is symmetrical about mean

It is asymptotic i.e. the curve never touches the horizontal axes

It is defined by the parameters; mean and standard deviation

The area under the normal curve is equal to 1

STANDARD NORMAL DISTRIBUTION (Z)

40 | P a g e
 X

Z =

OF STANDARD NORMAL TABLES

THEORM OF STANDARD NORMAL

P 0  Z  1.65

 1.65
= i.e. go and find 1.65 into the table under Z

= 0.4505

P 0  Z  1.96 

 1.96 
=

= 0.4750

P  1.24  Z  0 

 1.24 
=

= 0.3925

USE

P Z  1.65 0.5   1.65

=

= 0.5 + 0.4505

41 | P a g e
 = 0.9505

P Z  1.95

0.5   1.95
=

= 0.9744

P  0.95  Z  1.75

 0.95  1.75
=

= 0.3289 + 0.4599

= 0.7888

P 1.11  Z  2.54   2.54   1.11

=

= 0.4945 – 0.3665

= 0.128

P  3.58  Z  1.77   3.58  1.77 

=

P Z  1.96  0.5   1.96 

=

= 0.5 – 0.4750

= 0.025

P Z  2.68 0.5   2.68

=

= 0.5 – 0.4963

= 0.0037

EXAMPLES

42 | P a g e
A study of past participants of a certain program indicates that the mean length of time
spend on the program is normally distributed with a mean of 500Hrs and a standard
deviation of 100hrs

Find the probability that a candidate selected at random will:

Require between 500hrs and 580hrs to complete the program

Take less than 600hrs to complete the program

Take more than 700hrs

Take between 420hrs and 650hrs

Take between 410 and 490hrs

Solution

NB: mean = 500hrs, standard deviation = 100hrs

 500  500 580  500 

P Z 
P 500  X  580   100 100 
=

P 0  Z  0.8
=

 600  500 
P Z  
P X  600   100 
=

P Z  1
=

=
0.5  1

= 0.5 +

 700  500 
P X  
P X  700   100 
=

P Z  2 
=

=
0.5   2

43 | P a g e
  420  500 650  500 
P Z 
P 420  X  650   100 100 
=

P  0.8  Z  1.5
=

=
 0.8  1.5

 410  500 490  500 

P Z 
P 410  X  490   100 100 
=

P 0.9  Z  0.1
=

=
 0.9   0.1

= 0.3159 – 0.0398

= 0.2761

LINEAR PROGRAMING
STEPS IN CONSTRUCTING LINEAR PROGRAMMING

First identify the object of the question

The constraint i.e. the inequality

Stating the linear program modern

Then construct it on the graph

EXAMPLE

To start a new bus company, Mr X needs at least 5 buses and 10 minibuses. He does not
want more than 30 vehicles altogether. A bus takes up 3 units of space and a minibus takes
up 1 unit of a garage spaces and there are 54 units altogether. Running costs are $90 000 a
day for a bus and $48 000 a day for a minibus.

Determine graphically the maximum daily costs and corresponding number of buses and
minibuses. (14 marks)

Solutions

44 | P a g e
Objective function

To maximize the daily cost (C) = 90000x + 48000y

Let X being the number of buses

Let be Y being the number of minibuses

Subject to:
x  y  30 (Quantity)

3 x  y  54 (Capacity)

x5

y  10

VEHICLE TYPE QUANTITY CAPACITY NUMBER COST (C)

OF UNITS
Bus (x) 1 3 90 000
Minibus (y) 1 1 48 000
LIMITATIONS  30  54

THEN PLOT THE GRAPH

EXAMPLE 1

ABC Ltd makes two products, table and chairs which must be produced through Assembly
and Finishing department. Assembling department has 60hours available and Finishing
Department can handle up to 48hours of work. Manufacturing one table requires 4hours in
Assembling and 2hours in Finishing department. Each chair requires 2hours in Assembling
and 4hours in Finishing Department. If profit is $8 per table and $6 per chair, determine the
best possible combination of the number of table that can be produced in order to
maximize profits?

SOLUTION

Let be X being the number of tables

Let be Y being the number of chairs

45 | P a g e
Maximum Profits (P) =
8x  6 y

Subject to :
4 x  2 y  60 (Assembling Department)

2 x  4 y  48 (Finishing Department)

Constraint : x0

y0

PRODUCT TYPE ASSEMBLING FINISHING PROFIT (P)

DEPRTMENT DEPARTMENT
Tables (x) 4 2 $8
Chairs (y) 2 4 $6
LIMITATIONS  60  48

then construct a gragh on graph paper

FINANCIAL MATHEMATICS
Interest

We have two types of interest i.e. (simple and compound interest)

Annuities

We have two types of annuities i.e. (Present Value and Future Value of an annuity)

Simple interest

It is charged once a year and it is expressed over a given period. The interest is charged on
the principal once a year.

I = PRT

Key: I = Amount of interest earned in currency

P = The original principal borrowed

R = Interest rate per annum

46 | P a g e
 T = Time in year

FV = P(1+in)

Key FV = Future Value

P = Principal

I = Interest rate per annum

N = Time in years

COMPOUND INTEREST (SIMPLE INTEREST PLUS ACCRUED)

FV = P(1+in)

KEY FV = Future Value

P = Principal

I = interest per given period

N = number of period

EFFECTIVE RATE OF INTEREST (R) (r)

The actual interest charged per year

m
 i 
1    1
R =  m

Key r

I = Interest per year

M = number of compounds/periods per year.

EXAMPLE

Bank A pays interest of 18% compound annually, Bank B pays 17% compound monthly, Bank
C pays 16% compound daily. Which Bank pays the highest effective rate of interest?

m
 i 
1    1
Bank A: r =  m

47 | P a g e
 1
 0.18 
1   1
=  1 

= 0.18

= 18%

m
 i 
1    1
Bank B : r =  m

12
 0.17 
1   1
=  12 

= 0.1839

= 18.38%

m
 i 
1    1
Bank C : r =  m

365
 0.16 
1   1
=  365 

= 0.17346

= 17.35%

What rate of interest is required for $122 million to amount to $168 million in 8 months in
simple interest?

FV = P(1+in)

 8
1221  i 
168 =  12 

168 = 122 + 81.33i

168 – 122 = 81.333i

46 = 81.333i

ANNUITIES
48 | P a g e
They are associated with a fixed payment over a specified period of time. E.g. money for
investments and loan repayments

Types of annuities

Present value of an annuity (An)

1  1  i  
n

R 
An  i 
=

Key R = periodic payment

I = interest per given compound on period

N = number of compounding period (Total compound)

An
= present value

Example

Mr Dhuku borrowed $20 000 at 16% interest per year. Determine the regular payment, R if
interest is compounded:

Quarterly

Half yearly

NOTE: Duration of loan repayment is five (5) years

SOLUTIONS

1  1  i  
n

R 
An  i 
=

1  1  0.04 20 
R 
 0.04 
20 000 =

20 000 = R(13.59033)

R = $1 471.63

EXAMPLE 2

49 | P a g e
A person borrowed $10 000 at 15% per annum compounded half yearly over a period of
four (4) years. Calculate the periodic payment

SOLUTIONS

1  1  i  
n

An R
 i


=

1  1  0.0758 
R 
 0 .075 
10 000 =

10 000 = R(5.8573)

R = $1 707.27

AMORTISATION
It is associated with loan repayment

Amortisation schedule

Number Periodic Interest paid (i) Principal repaid Outstanding

payment (R) balance
0 - - - 10 000
1
2
.
n 0

Example1

A college wishes to borrow $50 000 to purchase a mini-bus. The debt is to be amortised at
an interest rate of 10% per annum for 8 years. Calculate:

The annual repayment (4)

Construct an amortisation schedule (14)

Find the outstanding principal after the 6th payment (2)

50 | P a g e
SOLUTIONS

1  1  i  
n

R 
An  i 
=

1  1  0.18 
R 
 0 .1 
50 000 =

50 000 = R(5.33492)

R = $9 372.20

NUMBER PERIODIC INTEREST PAID PRINCIPAL OUTSTANDING

PAYMENT (R) (i=0.1) REPAID BALANCE

0 - - - 50 000

1 9 372.20 5 000 4 372.20 45 627.80

2 9 372.20 4 562.78 4 809.42 40 818.38

3 9 372.20 4 081.84 5 290.36 35 528.02

4 9 372.20 3 552.80 5 819.40 29 708.62

5 9 372.20 2 970.86 6 401.34 23 307.28

6 9 372.20 2 330.73 7 041.47 16 265.81

7 9 372.20 1 626.58 7 745.62 8 520.19

9 372.21
8 852.02 8 520.19 0.00

The outstanding after the 6th is $16 265.81

51 | P a g e
Example 2

A debt of $60 000 is to be repaid through Amortisation over 2 years. The rate of interest is

12% per annum compounded quarterly. Calculate:

i. The quarterly payment applicable

ii. Draw up the amortisation schedule

iii. Indicate the capital and interest portion of 5th payment. (14)

SOLUTIONS

1  1  i  
n

R 
An  i 
=

1  1  0.038 
R 
 0 . 03 
60000 =

60000 = R(7.01969)

R = $8 547.38

NUMBER PERIODIC INTEREST PAID PRINCIPAL OUTSTANDING

PAYMENT (R) (i=0.03) REPAID BALANCE

0 - - - 60 000

52 | P a g e
1 8 547.38 1 800 6 747.38 53 252.62

2 8 547.38 1 597.58 6 949.80 46 302.82

3 8 547.38 1 389.08 7 158.30 39 144.52

4 8 547.38 1 174.34 7 373.04 31 771.48

5 8 547.38 953.14 7 594.24 24 177.24

6 8 547.38 725.32 7 822.06 16 355.18

7 8 547.38 490.66 8 056.72 8 298.46

8 8 547.41 248.95 8 298.46 0.00

At 5th payment the capital is $24 177.24 and the interest portion is $953.14

Sn 
FUTURE VALUE OF ANNUITY

It is associated with investments and it is deposited for a periodic payment.

 1  i n  1
Sn  R
 i


=

Sn 
Where = the future/maturity value

53 | P a g e
 R = periodic payment

I = interest per compound / period

N = number of compounds

EXAMPLE

What is the future value of annuity of $4 000 at 12% per annum compounded quarterly for

5 years. (4)

SOLUTIONS

 1  i n  1
Sn  R
 i


=

 1  0.0320  1
4000  
 0. 03 
=

= $107 481.50

Example 2

An investor deposit $1000 at the end of each year for five (5) years and earns interest at

15% per annum on her investment. How much will be accumulated in account at the end of

54 | P a g e
5 years? (4)

SOLUTIONS

 1  i n  1
Sn  R
 i


=

 1  0.155  1
1000  
 0 .15 
=

= $6 742.38

SINKING FUND

It shows number of periods, amount, interests, principal’s increase and investment value for
a future value of annuity.

NUMBER PERIODIC INTERST INCREASE IN BALANCE/

PAYMENT (R) EARNED (i) PRINCIPAL INVESTMENT

VALUE

0 - - - 0

55 | P a g e
Example 1

A firm wishes to accumulate a fund of $200 000 000 in 5 years by paying equal annual

investments earning 4% annually interest. Calculate:

Annual payment

Construct a sinking fund schedule for the accumulation of fund.

SOLUTIONS

 1  i n  1
R 
Sn   i 
=

 1  0.04 5  1
R 
 0 .04 
200 000 000 =

2000000000  5.41632256 
R 
5.41632256  5.41632256 
=

R = $36 925 422.70

NUMBER PERIODIC INTERST INCREASE IN BALANCE/

PAYMENT (R) EARNED (i) PRINCIPAL INVESTMENT

56 | P a g e
 VALUE

0 - - - 0
36 925 422.70 36 925 422.70 36 925 422.70
1 0

2 36 925 422.70 1 477 016.91 38 402 439.61 75 327 862.31

3 36 925 422.70 3 013 114.49 39 938 537.19 115 266 399.50

4 36 925 422.70 4 610 655.98 41 536 078.68 156 802 478.20

36 925 422.70
5 6 272 099.13 43 197 521.83 200 000 000.00

SAMPLING THEORY
Confidence interval (C.I)

It is the interval or range of value in which we inspect to find the population parameter at a

specified percentage level (Confidence level)


Confidence interval for the true population mean,

SITUATIONS


i.e. sigima known

Sigima is the number from the population


 Z.
x n

57 | P a g e
KEY

x
= sample mean

Z
= normal distribution table figure

N = is the number of the population

 n  30
(sigima) is unknown and n is large

S
X  Z.
n
=

S is the sample standard deviation

 n  30
(sigima) unknown and n is small

S
X  .t 
2
n  1.
n
=

Key

T = is the use of distribution tables

Confidence interval C.I for the true population proportion

58 | P a g e
 PQ
P  Z.
n
=

EXAMPLE1

After a particular wet night, 12 worms surfaced on the lawn, the length measured were in

centimetre were:

9,5 9,9 11,2 10,6 9,5 11,1 10,9 9,8 10,1 10,2 10,9 11

Then construct a 95% confidence interval for the true population mean of worms. (8)

SOLUTIONS

x
x = n
Mean,

9,5  9,9  ...  11

12
=

= 10,3917 (4dp)

 x2 n x 
2

n 1
Variance =

59 | P a g e
  x2 n x 
2

n 1
Standard deviation =

1300 , 2312 10 , 39167 2

121
=

= 0,6317 (4dp)

S
X  .t 
2
n  1.
n


= 1 – 0,95 = 0,05

t(11) = 2,201

0,63
10,39  2,2 
12
=

10,39  0,4003
=

= (10,39 – 0,4003) (10,39 + 0,4003)

= (9,9897 ; 10,7903)

60 | P a g e
EXAMPLE 2

The weight of 7 similar boxes in kg of dog food is as follows: 9,8 10,4 10,0 9,0

10,2 9,6 9,9. Find the 95% confidence interval for the mean of all dog foods.

Assume an approximately normal distribution.

SOLUTIONS

S
X  .t 
2
n  1.
n

x
mean, , , x n
=

9,8  10,4  ...  9,9

7
=

= 9,8429

 x2 n x 
2

n 1
Standard deviation =

679 , 417 9 ,8429 2

7 1
=

= 0,4541 (4dp)

61 | P a g e
Therefore, alpha = 1 – 0,95 = 0,05

t(7) = 2,365

S
X  .t 
2
n  1.
n

0,45
9,84  2,365 
7
=

9,84  0,4022
=

= (9,84 – 0,4022) (9,84 + 0,4022)

= (9,4378 ; 10,2422)

EXAMPLE 3

The sales manager of a firm that manufactures light bulbs carried out an investigation on

the life average life of these light bulbs. A sample of 36 these bulbs showed a mean of

1430hours with a standard deviation of 100hours. Construct a 95% confidence interval for

the true average life of light bulbs.

Solutions

62 | P a g e
 S
X  Z.
n

Draw a diagram

0,5 – 0,025 = 0,475

 Z   0,475

Z  1,96

S
X  Z.
n

100
1430  1,96 
36
=

1430  32,66667
=

= (1430 – 32,66667) (1430 + 32,66667)

= (1397,33 ; 1462,67)

EXAMPLE 4

A sample of 100 people was selected at a tax conference. Their ages were recorded and the

mean age was 32 years with a standard deviation of 6 years. Construct:

63 | P a g e
 i. 98% confidence interval for the mean age

ii. 99% confidence interval for the mean age

iii. 90% confidence interval for the mean age

iv. 96% confidence interval for the mean age

SOLUTIONS

Draw a 98% normal distribution

0,5 – 0,01 = 0,49

 Z   0,49

Z  2,33

S
X  Z.
n

6
32  2,33 
100
=

32  1,398
=

= (32 – 1,398) (32 + 1,398)

= (30,60 ; 33,398)

64 | P a g e
 EXAMPLE 5

A researcher wishes to investigate the proportion of people who regularly travel by a train.

Of a sample of 400 people, 285 said they did so. Construct a population confidence interval

at the following levels:

i. 99,5%

ii. 99%

iii. 94%

iv. 88%

v. 92,5%

SOLUTIONS

PQ
P  Z.
n

P  population ~ 0 f ~ success

285
P  0,7125
400

q 1 p

q  1  0,7125  0,2875

65 | P a g e
THEN DRAW A NORMAL DISTRIBUTION

 Z   0,4975

Z  2,81

PQ
P  Z.
n

0,7125  0,2875
0,7125  2,81 
400
=

0,7125  0,06359
=

= (0,6489 ; 0,7761)

= (64,89% ; 77,61%)

HYPOTHESIS TESTING

Hypothesis= it is a claim or statement/assumption that is brought forward and that need to

be validated.

Hypothesis test = it is a statistical procedure which is used to assess whether the given

66 | P a g e
claim/assertion is true or false.

Terms used in hypothesis test:

Null hypothesis (Ho)

It is the actual claim of assertion that needs to be validated.

Alternative hypothesis (H1)

It is an option to the new hypothesis null.

Critical value (CV)

This is the value obtained from statistical table which separate the rejection region from

acceptance region.

Test statistics

It is a value calculated from experimented data which helps one to go either for rejected or

non-rejected region.

HYPOTHESIS TESTING – PROCEDURE

1. State Ho and H1 or claim

2. State the level of Test/Significance level

3. Identify the nature of distribution involved i.e. normal or t tables

4. Determine/calculate the critical value (CV) and hence state the rejection criteria

5. Calculate the test statistics

67 | P a g e
 6. Compare 4 and 5 and hence draw up your conclusion

HYPOTHESIS TESTING FOR THE TRUE POPULATION MEAN

Situations

If the sigima is known we use Z-test

n  30
, we use Z-test
If the sigima is unknown and n is large

n  30
If the sigma is unknown and n is small , we use T-test

POSSIBLE HYPOTHESIS (STEP 1)

 
1. Ho : = 60

  
H1 : 60 (if it is one tailed)

 
2. Ho : = 60

  
H1 : 60 (if it is one tailed)

 
3. Ho : = 60

  
H1 : 60 (if it is two tailed)

68 | P a g e
Stating or examples of hypothesis

It is thought that the mean time of a person uses a credit card in a year is 180 neither more

or less. To test this hypothesis, a sample mean was found to be 192 with a standard

deviation of 50. What conclusion will be reached at a 5% significance level? (10)

SOLUTION


= 180 (180 is the population mean)
Ho :


H1 : 180 (it is two tailed)

Test at 5% significance level

Sigma is unknown and n is large, we use Z-test

The critical value (Z) = 1,96

Then draw a diagram and lable the rejected and accepted region

x 
s/ n
Z calc =

192  180
50 / 55 
=

= 1,7799 (4dp)

Then compare 4 and 5 and hence conclude it

As Z calc = 1,7799 and in the acceptance region, we do not reject Ho and concluded that at
5% significance level that the mean time of a person uses a credit card is 180.

Example 2

69 | P a g e
A trade union claims that the average hourly rate paid to domestic clean is only $2.85. A
house league wants to test this claim. The league therefore conducted a survey amongst a
250 domestic cleaners throughout the country and found that the hourly paid for cleaners
was $3.03 with a variance of $1. Test the hypothesis at 5% significance level that the
average hourly paid rate to domestic cleaners is indeed $2.85. (18 marks)

SOLUTION


= 2.85
Ho :


H1 : 2.85 (2 tailed)

Test at 5% significance level

Sigma is unknown and n is large, we use Z-test

The critical value (Z) = 1.96

Then draw the diagram and show the accepted and rejected region

x 
s/ n
Z calc =

3.03  2.85
1 / 250
=

= 2.846 (3 dp)

As Z calc = 2.846 and in the rejection region, we reject Ho and conclude that the mean

average hourly rate paid to domestic cleaners is not equal to $2.85.

70 | P a g e
Example 3

At a forum of the Deans students held last year, most delegates held a view that college

students washed average of 6.8 videos monthly. A random sample of 36 college students

reviewed that the mean number of videos washed last month was 6.2 with a standard

deviation of 0.5. Is there any evidence that students wash fewer videos? Test at 5%

significant level. (15)

SOLUTION


= 6.8
Ho :

H1 : < 6.8 (one tailed)

Test at 5% significance level

Sigma unknown and n is large ~ we use Z – test

CV (Z) = 1.65

THEN DRAW A DIAGRAM AND SHOW THE ACCEPTED AND NON-ACCEPTED REGION

x 
s/ n
Z calc =

6.2  6.8
0.5 / 36
=

71 | P a g e
 = -7.2

As Z-calc = -7.2 and we reject Ho and concluded that students watched less than 6,8 videos

per month.

EXAMPLE 4

It is thought that the mean net weight of bags of sugar produced by a machine is at least

1.03kg. a sample of 230 bag had a mean net weight of 1.02kg with a standard deviation of

0.02kg. test the hypothesis mean is at least 1.03kg at the 1% significance level. (12)

SOLUTION

 
1.03
Ho :


H1 : < 1.03 (one tailed)

Test at 1% significance level

Sigma is unknown and n is large ~ we use Z-test

Critical value (Z) = 2.33

THEN DRAW THE DIAGRAM AND LABEL THE ACCEPTED AND NON-ACCEPTED REGION

72 | P a g e
 x 
s/ n
Z calc =

1,02  1.03
0.02 / 230
=

= -7.58 (2 dp)

As Z-calc = -7.58 and we reject Ho and concluded that the mean net weight of bags of sugar

produced by a machine is less than 1.03kg.

73 | P a g e