Professional Documents
Culture Documents
QUANTITATIVE TECHNIQUES
COURSE OUTLINE
QUANTITATIVE TECHNIQUES
Purpose: To equip students with skills in quantitative techniques necessary for decision making
in business
Expected Learning Outcomes of the Course: By the end of the course unit the learners should
be able to:-
i. Apply sets theory, probability and linear programming in decision making
ii. Summarize raw data into information for decision making
Course Content:
Measures of central tendencies; Measures of dispersion; Probability- Axiom definition of
probability;; Conditional probability; Probability trees; Linear equations, inequalities and their
applications; Utility functions and curves
Course Assessment
Examination - 70%; Continuous Assessment Test (CATS) - 20%; Assignments - 10%; Total -
100%
Recommended Text Books:
Sweeny Williams Anderson (2007), Quantitative Methods For Business, McGraw Hill,
New York
Oakshott Les (2006), Essential Quantitative Methods For Business, Management And
Finance Routledge, London
B S Sharma(2006); Quantitative Methods; Anmol Publications Pvt
Anderson (2007); Quantitative Methods; Cengage Learning ( Thompson )
Louise Swift and Sally Piff (2005), Quantitative Methods For Business, Management
And Finance McGraw Hill, New York
TOPICCS SUB-TOPICS
WK
Median
Mode
Weighted mean
Geometric mean
Harmonic mean
Operational Research is the attack of modern science on complex problems arising in the
direction and management of large systems of men, machines, materials and money, in industry,
business, government and defense, the distinctive approach is to develop a scientific model of the
systems incorporating measurements of factors such as change and risks, with which to predict
and compare the outcome of alternative decisions, strategies or controls. The purpose is to help
management determine its policy and action scientifically.
The above formal definition contains several essential features and these are;
The basis of the OR approach is that of constructing models of problems in an objective, factual
manner and experimenting with these models to show the results of the various possible course
of action. A model is any representation of reality and may be in graphical, physical or
mathematical terms. The type of model most frequently used in OR is a mathematical model,
i.e.one which tries to show the working of the real world by means of mathematical symbols,
equations and formulae.
Y= a + bx
Is a model of the relationship of total overheads to the number of units produced where?
Y = Total overheads
The above model is obviously very simple and most practical models are of necessity much more
complex. It is important to realize that however complex looking a model is and however many
variables it contains, it still involves considerable simplification of reality and any results or
predictions obtained from the model must therefore be used with caution and judgment.
Although mathematical or symbolic models are common, other types of models e.g. Iconic,
Iconic models
They are visual models of real object(s) they represent. They may be larger or smaller than the
reality. For example, a model steam engine is much smaller than the real thing whilst the
familiarcolored plastic models of molecular structures are much larger. Both of these iconic
models and so are pictures, maps and diagrams although these latter are in a different form to the
reality they present. These models are often difficult to manipulate experimentally (but not
always e.g. wind tunnel model for aircraft) so are not used greatly in operational research.
Analogue model
They use one set of physical movements or properties to represent another set. For example, the
movement of a piece of metal under stress can be represented in a more observable form by the
movement of a gauge finger. Fathom lines on charts and lines on graphs are also analogues of the
reality they represent.
Although analogue models are more versatile than iconic models they generally lack the
flexibility of mathematical models.
Simulation models
It represents the behavior of a real system. Perhaps the best known example is the flight
simulator used in the training of pilots. Simulation models are used where there is no suitable
mathematical model, where the mathematical model is too complex, or where it is not possible to
experiment upon a working system without causing serious disruption – as in the case of the
training of pilots. One application of simulation models in the management context is the study
of the behavior of people and objects in queues. With the development of relatively inexpensive
computing facilities many business games are based on the stimulation of the operation of
complete business.
Heuristic models
These are models which use a set of intuitive rules which managers hope will produce at least a
workable solution, and a better solution than methods currently being used. For example, a
deliver van driver may be instructed to plan the day’s deliveries using the following rule, after
each call delivery to the nearest customer whom you have not yet visited. This will certainly give
a good solution early in the day, but it can lead to some long distance being travelled at the end
of the day back to the depot. The driver has no way of knowing whether the route gives optimum
time and distances and any improvements can come only through testing other heuristic
approaches.
Models may be further classified into normative and descriptive. Normative models are
concerned with finding the best, optimum or ideal solution to a problem. Many mathematical
models fall into this classification. Descriptive models, as their name implies, describe
thebehaviour of a system without attempting to find the best solution to any problem. For
example, simulation tends to fall within this category.
Perhaps the most important point to appreciate is not so much what a model is called but what it
does in helping managers to attain the goals that they have set.
The primary aim of OR is to attempt to identify the best way of conducting the affairs of the
organization i.e. the optimum. In studying problems the OR practitioner tries to optimize the
operation of the organization as a whole rather than narrow aspects of the business such as a
single department or section. This is easier said than done and because of the practical necessity
of dealing with manageable areas of work and thereby producing simplified and incomplete
models of operations there may be a tendency to produce suboptimal solutions, i.e. a solution
which is optimal for a small section of the firm, but not optimal for the firm as a whole. This is
another point which should be watched when considering the results of an OR investigation.
All business planning and decision making involves forecasting future activities. This cannot be
done with any certainty and so to provide the maximum possible assistance to planners and
decision makers a systematic analysis of the possible extent of the risks and uncertainties
involved is a vital part of any OR study. OR techniques do not of themselves remove the risks
and uncertainties, but they are able to highlight their effects on the firms operations. An
important method to help with this process is what is termed sensitivity analysis. When a
solution has been obtained using one of the OR techniques ( such as linear programming,
inventory control, investment appraisal) alterations are made to the factors of the problem, such
as , sales, costs, amount of materials, to see what effect there is on the original solution. If the
value of the original solution alters considerably with minor changes in the factor values it is said
to be sensitive. In such circumstances the whole problem will need much deeper analysis and
particularly sensitive factor may cause decisions to be altered.
In general OR practitioners do not make the business decisions. Their role is the provision of
information to assist the planners and decision makers. The skill, experience and judgment of
managers cannot be replaced by formal decision making techniques. The results of an OR
investigation are but one input of information into the decision making process. There is a strong
parallel between the OR practitioner supplying information for management decision making
and that of other information specialists such as accountants. This is why a knowledge of OR for
the accountant (and accountancy for the OR practitioner) can be very useful and can improve
quality of the information provided. There are strong reasons why the most effective OR teams
contain people drawn from various background –economists, accountants, mathematicians,
engineers, psychologists etc. in this way there is more chance that the numerous facets of
business problems can be recognized and analyzed.
Quantification of factors
Not all factors involved in a decision making or planning situation can be quantified, but themost
readily usable OR techniques are those based on quantifiable factors such as costs,
revenues, number of units etc. these techniques are the ones most commonly included in
examinations and form the basis of the syllabuses of the major professional bodies, hence the
contents and title of this book.
Revision Exercise
5. Give reasons why the results of an operational research study may not be implemented?
An equation is a mathematical statement that relates two algebraic expressions involving at least
one variable.
Properties of equality
5. Substitution property: If a=b, then either may replace the other in any statement without
changing the truth or falsity of the statement.
(a) Unknown
These are the variables whose values are not given. They are found by solving the equation e.g.
in the equation 4x + 5y =40 and are unknowns. x and y .
(b) Constants
They are fixed figures, which are shown on the left hand side of unknowns or separately. From
the equation 4x + 5y =40, 4, 5 and 40 are constants.
(c) Coefficients
They are those figures, which are shown on the left hand side of unknowns. From the equation
4x + 5y =40 , 4 and 5 are coefficients of and respectively.
(d) Index
It is the power of an unknown. It is shown on the right hand side on top of any unknown e.g. in
Variables
Definition of a variable
It is a measurable characteristic that assumes different values among the subjects.
Types of variables
1. Independent variables / Predictor variables
Experimental variables: They are variables which the researcher has manipulative control over
them. Are commonly used in biological and physical sciences e.g. influence of amount of
fertilizer on the yield of wheat, influence of alcohol on reaction time.
ii. Measurement types of independent variables: Are variables, which have already occurred.
They have fixed manipulative and uninfluenceable properties. Most of the variables are either
environmental or personal e.g. age, gender, marital status, race, colour, geographical location,
nationality, soil type, altitude etc. (e.g. influence of nationality on choice of food).
They attempt to indicate the total influence arising from the effects of the independent variable.
It varies as a function of the independent variable e.g. influence of hours studied on performance
in a statistical test, influence of distance from the supply center on cost of building materials.
3. Discrete variables
They are variables, which can only assume whole numbers. They are always counted e.g.
number of cars, children, trees books etc.
4. Continuous variables
They are variables, which can assume any value in a specific range. They are always measured
e.g. height, weight, weight, radius, distance etc.
5. Extraneous variables
They are those variables that affect the outcome of a research study either because the researcher
is not aware of their existence or if the researcher is aware, she or he has no control over them.
They are extraneous variables that are built into the study. They are introduced to increases the
validity of the data and to leads to more convincing generalizations.
Types of equations
They are equations with one unknown and the index of the unknown is one. It is of the form ax +
b =0. Where a and b are constants and the solution is obtained by writing ax= -b
x - 4 = 10
Solution
x = 14
The answer is x = 14
Check the solution by substituting 14 in the original equation for x. If the left side of the equation
equals the right side of the equation after the substitution, you have found the correct answer.
2x-4=10
=2x=10+4
=2x=14
X=14/2
X=7
5x-6=3x–8
6x-7 +3x-5
4 7
2(3x - 7) + 4 (3 x + 2) = 6 (5 x + 9) + 3
Complete multiplication.
6x-14+12x+8=30x+54+3
18x-6=30x+57
-6=12x+57
-63=12x
X=21/4
6x - 6 > 2x + 2
Solution
Given 6x - 6 > 2x + 2
Conclusion The solution set consists of all real numbers in the interval (2 , + infinity).
Solution
Solution
Multiply factors and group like terms -3 <4x + 8 - 3 < 9 -3 < 4x + 5 < 9
Subtract 5 to all three terms and simplify -3 - 5 <4x + 5 - 5 < 9 - 5 -8 < 4x < 4
Divide all three terms by 4 -2 < x < 1
Conclusion The solution set consists of all real numbers in the interval (- 2 , 1)
Essentially you graph the boundary line the same as if the problem was a linear equation.
Assume that there is an equal sign and use an appropriate method to graph the line. Unless the
directions to a problem indicate otherwise, you can use any method to graph it. Two ways of
graphing linear inequalities are
plotting any three points or
Using the x and y intercepts.
When you draw the boundary line, you need to have a way to indicate if the line is included or
not in the final answer.
For a solid boundary line, < or >are used in the inequality. Otherwise the dashed line is used.
The graph below shows the boundary line for x + y < 6: (note that this does not show the
inequality part)
Revision Exercise
i) y x + 3
ii) y > 2x – 1
iii) 2y 4x + 6
iv) 2x + y > 4
Introduction
Definition
It is a mathematical technique that deals with the optimization of a linear function of variables
known as objective function to a set of linear inequalities known as constraints.
ii. The constraint set: It is a system of equalities and / or inequalities describing the conditions
under which optimization is to be accomplished e.g. machine hours, man-hours, materials etc.
Divisibility: Quantities, revenues and costs are infinitely divisible i.e. any fraction or
decimal answer is valid.
Certainty: The technique makes no allowance for uncertainty in the estimate made.
Costs and benefits which cannot be quantified easily, such as liquidity, good will and
labour stability are ignored.
It takes into account single objective only i.e. profit maximization or cost minimization
It can be adopted only under the condition of certainty i.e. that the resources available,
per unit contribution, costs etc are known with certainty. This does not hold in real
situations.
Linear programming can find use in many areas where decisions need to be made based on
scarcity of recourses. A few areas are mentioned below:.
Paper and textile industry: To determine the optimal cutting method to minimize trim
losses
If there are n decision variables and m constraints in the problem, the mathematical formulation
of the LP is:-
a x +a x +…a x <b
11 1 12 2 1n n 1
a x +a x +…a x <b
21 1 22 2 2n n
a x +a x +…a x <b
m1 1 m2 2 mn n m
Where
x - decision variable –
i
th
a constant representing exchange coefficient of the j decision variable in the
ij- -
th
i constraint
th
b constant representing i constraint requirement or availability
i---
Solutions to LP problems
i. Graphical method
ii. Graph the constraint inequalities as follows: treat each inequality as though it were an
equality and for each equation, arbitrarily select two sets of points. Plot each of the points
and connect them with appropriate lines.
iii. Identify the solution space or feasible region which satisfies all the constraints
simultaneously. For ≤ constraints, this region is below the lines and for ≥ constraints; the
region is above the lines.
iv. Locate the solution points on the feasible region. These points always occur at the corner
points of the feasible region.
Example
At the start of the current week there are 30 units of X and 90 units of Y in stock. Available
processing time on machine A is forecast to be 40 hours and on machine B is forecast to be 35
hours.
The demand for X in the current week is forecast to be 75 units and for Y is forecast to be 95
units. Company policy is to maximize the combined sum of the units of X and the units of Y in
stock at the end of the week.
Formulate the problem of deciding how much of each product to make in the current week as a
linear program.
Solve this linear program graphically.
Solution
Let
y be the number of units of Y produced in the current week then the constraints are:
The objective is: maximize (x+30-75) + (y+90-95) = (x+y-50) i.e. to maximize the number of
units left in stock at the end of the week
It is plain from the diagram below that the maximum occurs at the intersection of x=45 and 50x
+ 24y = 2400
Revision exercise
1. A firm is engaged in producing two products A and B. each unit of product A requires 2 kg of
raw material and 4 labour hours for processing, whereas each unit of product B requires 3 kg
of raw material and 3 hours of labour, of the same type. Every week, the firm has an
availability of 60 kg of raw material and 96 labour hours. One unit of product A sold yields
Sh. 40 and one unit of product sold gives Sh. 35 as profit. Formulate this problem as a linear
programming problem to determine as how many units of each of the products should be
produced per week so that the firm can earn the maximum profit. Assume that there is no
marketing constraint so that all that is produced can be sold.
2. The Kenya Agricultural Research Institute suggested to a farmer to spread out at least 4800 kg
of a special phosphate fertilizer and not less than 7200 kg of a nitrogen fertilizer to raise
productivity of crops in his fields. There are two sources for obtaining these mixtures A and B.
Both of these are available in bags weighing 100 kg each and they cost 115 Sh. 40 and Sh 24
respectively. Mixture A contains phosphate and nitrogen equivalent of 20 kg and 80 kg
respectively, while mixture B contains these ingredients equivalent of 50 kg each. Determine
how many bags of each type the farmer should buy in order to obtain the required fertilizer at
minimum cost.
Invert a Matrix
Presentation of information
Accountants and businessmen are well versed with presenting information in tabular form or
writing information in rows and columns. Often one set of information in tabular form e.g.
depreciation by category of fixed asset, is worked upon to produce a subsequent set of
information based upon the first table, e.g. depreciation by category of fixed asset spread over
cost centre’s. Although not generally referred to as such the above are examples of matrices. An
understanding of matrices and matrix algebra may provide short cut methods of calculation and
will provide further insights into improved methods of presenting and manipulating data.
Matrix definition
A matrix is a rectangular array of numbers whose value and positioning in the matrix is
significant. A matrix is usually, but not always, shown in brackets thus,
18
37
Example 1
The size of a matrix is given by the number of rows and the number of columns, i.e. rows x
columns. The symbols most commonly used being m (columns) x n (rows). The matrix in
Example 1 is a 2 x 2 matrix, and as the number of columns equals the number or rows, it is
known as square matrix.
The following are further examples matrices:
12
3 4(3 x 2 matrix)
56
5 9 16
2 8 4(2x 3 matrix)
8 12 6
951 (3x 3 matrix i.e. squire
702
Matrix algebra
The particular rules applying the manipulation of data in matrix form are given in the following
sections on matrix addition, matrix addition, matrix subtraction, matrix multiplication and matrix
inversion. As a form of shorthand matrices are often referred to by capital letters, for example
A =1 6
52
83
C=32
16
Matrix addition
The only rule is that matrices to be added (or subtracted) must be the same size as one another,
I.e. they must have the same number of columns and the same number of rows, for example.
Example
A= 1 11 2
629
B= 2 0 7
596
1 11 2 2 0 72 11 9
A+B= +=
629 5 9 611 11 15
1 2 11 0 2 7 1 2 18
Matrix subtraction
Matrix subtraction uses the same general rules as matrix addition. For example, using the same
matrices as in
Example 8 calculate
1 11 2 2 0 7-1 9 -5
A-B - =
6 2 95 9 61 -7 3
Matrix multiplication
There are two aspects of matrix multiplication, the multiplication of a matrix by a single number,
called a scalar, and the multiplication of a matrix by another matrix.
Scalar multiplication
A scalar is an ordinary number such as 3,6,8,2 etc. The rule for this is simply-multiply each
element in the matrix by the scalar, for example.
Let A =5 2
83
4xA= 4x5 2 = 20 8
8 3 16 12
Multiplication of matrices
A= 3404 28 52
1 -1 710 = -7 -6
2 32 1 38
32 -2 = 6
42 3 -2
3 101 = 303
1 101
-7 -7 0 -7
cd
A =ad-cb
Example
B 3 1 B =3X4- 21 =10
24
-2/10 3/10 24 01
Revision Exercise
1. How is the size of a matrix described?
2. What is a vector?
5. What is scalar multiplication?
6. What is the size rule for matrix multiplication?
7. How is matrix multiplication carried out?
8. What is the unity matrix?
9. Does A x A-1 = A-1 x A?
10. What are the steps in matrix inversion?
11. What are the essential conditions for simultaneous equations to be solved by matrix algebra?
12. What are probability transition matrices and what are they used for?
13.
X= 3 11 6
938
509
Y= 120
045
587
Calculate a) X+Y
b) X-Y
14. Multiply the matrices X and Y given in question 13
Cost
b fixed
Activities
Figure 1.Graph of a simple linear cost function
In such a function the rate of change of cost (represented by ‘b’, the gradient of the line) is
constant at all levels of activity and will not increase or decrease at any level of activity. This is,
of course, is what is meant by a linear function. The value of b can be easily found by simple
arithmetic without recourse to more sophisticated techniques.
Differentiation
The process of differentiation establishes the slope of a function at particular point. Alternatively
this can be described as establishing the rate of change of the dependent variable (say, cost) with
respect to an infinitesimally small increment in the value of the independent variable (say,
activity).
For illustration consider the following two functions:
Function I y = x
Function II y = x2
Function I is a linear function whereas, because x is raised to the power of 2 (i.e. x2) in
Function II, that function is curvy-linear.
Assume now that the independent variable, x, is altered by a very small amount, m.
What is the rate of change in y caused by the change in the value of x?
Function I
If x is at some value A, then y is also at the value A. If x is altered by m then its value becomes x
= A + m and, as a direct consequences, the value of the dependent variable becomes
Y=A+m
It follows therefore that:
The rate of change of y with x = change in value of y = m =1
Change in value of x m
Generally written as
Dy
Dx, which means the derivative of a function when x tends towards zero
Thus for the original function: y = x2
Derivative 2
Dy=2x
Dx
b) The derivative of a function gives the exact rate of change at a point and only gives
approximate results when used over a finite range. An example of the effect of this is where
differentiation is used to obtain the marginal cost from a curvy-linear cost function.
Accountants frequently define marginal cost as the increase in total cost due to an increase in
output of one unit. A whole unit is a finite range as far as differentiation is concerned and
consequently the marginal cost obtained from the derivative is only an approximation. This is a
technical point which is considered not to be of great practical significance
Dy =nx^n-1
Dx
Examples
Function Derivative
Y = 3x^2 = dy/dx=6x
Y = 8x^4 = dy/dx=32x
Y = -7x^5 =dy/dx=-35x
Application of differentiation
A practical example of differentiation
Now that the idea of differentiation has been explained and the rules given for differentiating
common functions, a practical example can be considered.
Illustration:
A firm has analyzed their operating conditions, prices and costs and has developed the following
functions
Revenue Ksh (R) = 400Q-4Q2 and Cost Ksh (C) = Q2+10Q+30
Where Q is the number of units sold
The farm wishes to maximize profit and wishes to know
i. What quantity should be sold?
ii. At what price?
Note: Previous examples have used y and x which results in the derivative
Dy/dx
This example uses R, C and results in the derivatives
dR/dQand .dC/dQ
Solution
From basic economic theory it will be recalled that profit is maximized when
Marginal Cost = Marginal Revenue and, as explained in this chapter, differentiating a function
gives the rate of change of that function which is equivalent to the marginal cost or revenue. Ksh
R = 400Q-4Q2 and dR/dQ= 400-8Q = Marginal Revenueand Ksh C = Q2+10Q+30 and
DC/dQ = 2Q+10 = Marginal Cost
Points of profit maximization is when
MR = MC ordRdC/=dQ/ dQ
I.e. 400-8Q = 2Q+10
Q = 39 answer (a)
Integration
For our purposes integration can be regarded as the reverse of differentiation. Differentiation
establishes the slope of the function at a point whereas integration can be defined as the
procedure for finding the area under the curve of a function i.e. integration is a process of
summation.
As integration is the reverse of differentiation
However the integral above is not complete because it will be recalled from the paragraphs that
X^4 + 10, or x^4 + 50, or x^4 + c, where c is any constant also have as their derivative 4x3 so
that is essential to recognize this possibility by writing
4x^3dx x^4 c
The integral including the undetermined constant is known as the indefinite integral. The value
of the constant can, in some instances, be inferred to be zero or it may have a value when
additional information known as an initial condition is supplied.
Examples
Functions Integral
Y=10x4 10x^4dx=10x^5 +c =2x^5+c
5
Notes:
a) It will be seen that a constant is added to each integral
b) Since differentiation is a more straight forward process is useful to check the integral by
differentiating it to see that it comes back to the original expression.
For instance differentiating 2x^5+c gives 10x4 which is the first example above
Revision questions
1. A cost function is
Ksh (c) = Q2-30Q+200
Where Q = quantity of units produced
Find the point of minimum cost.
5. A firm sells each unit of its product for Ksh 400. The cost function which describes the total
cost C as a function of the number of units produced and sold x is:
C = 40x + 0.25x2 + 250
Determine the maximum profit and the corresponding total revenue and total cost.
6. A government financial advisor has projected that the, cost of retrenchment compensation will
be at a rate of: r (t) = 10e0.5t million shillings per year t years from now. The government has
budgeted a maximum of
Ksh 300 million for retrenchment compensation fund.
When will this fund he exhausted.
7. A firm’s demand function is given as P=24 - 3x where p = price and x quantity produced and
sold.
Determine the output for maximum revenue & show it is a maximum. What’s the price at
maximum revenue?
8. The revenue function of a product is R= 28q - q2 and the unit variable cost v = q – 8 while
fixed cost is Sh 64
Required
(i)Total cost
(ii) Profit function
(iii)Output and price for maximum profit
(iv)Show that the output for maximum profit is not necessarily the same as the output for
maximum revenue.
Describe the theory behind mutually exclusive and non-mutually exclusive events and
classify experiments accordingly.
Evaluate the probability of mutually exclusive events by applying the addition rule.
Introduction
In the study of chance, we need a mathematical method to describe the likelihood of events
happening.
Probability is the study of chance (likelihood) of events happening.
The events are also referred to as outcomes. The study of probability has vitally important
applications in physical and biological sciences, economics, politics, sport, insurance, quality
control, planning in industry, and a host of other areas.
The range of probability
Probability is studied by assigning a number which lies between 0 and 1.
An event which a 0% chance of happening (“It did not rain on Tuesday” i.e., it is impossible) is
assigned a probability of 0.
An event which has a 100% chance of happening is assigned a probability of 1
All other events can then be assigned a probability between 0 and 1.
Definitions
Probability Experiment
Process which leads to well-defined results called outcomes
Outcome
The result of a single trial of a probability experiment .For example, in tossing of a coin, a head
or a tail constitute the two possible outcomes.
Sample Space
Set of all possible outcomes of a probability experiment It is denoted by ‘S’ and its number of
elements are n(s). For example; in throwing a dice, the number that appears on top is any one of
1, 2, 3,4,5,6. So here,
S = {1, 2, 3, 4, 5, 6} and n(s) = 6
Event
One or more outcomes of a probability experiment. In other words every subset of a sample
space is called an event and is denoted by ‘E’.
Example; In throwing a dice S = {1, 2, 3, 4, 5, 6}, the appearance of an even number will be the
event E = {2, 4, 6}.
Clearly, E is a subset of S.
Classical Probability
Uses the sample space to determine the numerical probability that an event will happen; also
called theoretical probability
Equally Likely Events
Events which have the same probability of occurring; for example, in tossing of a fair coin, the
appearance of a head or a tail are equally likely events.
Complement of an Event
All the events in the sample space except the given events.
Empirical Probability
Uses a frequency distribution to determine the numerical probability; an empirical probability is
a relative frequency.
Subjective Probability
Uses probability values based on an educated guess or estimate; it employs opinions and inexact
information.
Mutually Exclusive Events
Two events which cannot happen at the same time, for example in tossing a coin it is not
possible to obtain a head and a tail in a single toss. Similarly, events such as “It rained on
Tuesday” and “It did not rain on Tuesday” are mutually exclusive events. When calculating the
probabilities of mutually exclusive events you add up the probabilities.
Disjoint Events
Another name for mutually exclusive events
Independent Events
Two events are independent if the occurrence of one does not affect the probability of the other
occurring. For example the events “It rained on Tuesday” and “My chair broke at work” are two
independent events. When calculating the probabilities of independent events you multiply the
probabilities. If a coin is tossed twice, the two outcomes will be unrelated obtaining a head on
the first toss does not give a bearing on what will show up in the next toss.
Dependent Events
Two events are dependent if the first event affects the outcome or occurrence of the second event
in a way the probability is changed.
Conditional Probability
The probability of an event occurring given that another event has already occurred.
Bayes' Theorem
A formula which allows one to find the probability that an event occurred as the result of a
previous event
Axiomatic Approach to Probability
The modern theory of probability is based on the axiomatic approach introduced by the
Russianmathematician A.N Kolmogorov in 1930’s. In axiomatic approach, some concepts are
laid downat certain properties or postulates commonly known as axioms are defined and from
these axioms the entire theory is developed by logic of deduction.
For example, throwing a die once can yield a 5 or 6, but not both, in the same toss. The
probability that either a 5 or 6 occurs is the sum of their individual probabilities.
P=p1+p2
=1/6+1/6
=1/3
Illustration: From a bag containing 5 white balls, 2 black balls, and 11 red balls, 1 ball is drawn.
What is the probability that it is either black or red?
Solution: The draw can be made in 18 ways. The choices are 2 black balls and 11 red balls,
which are favorable, or a total of 13 favorable choices. Then, the probability of success is
p=13/18
Since drawing a red ball excludes the drawing of a black ball, and vice versa, the two events are
mutually exclusive; so the probability of drawing a black ball is
p1=2/18and the probability of drawing a red ball is
p2=11/18
Therefore, the probability of success is
P1+p2
=2/18+11/18
=13/18
Illustration What is the probability of drawing either a king, a queen, or a jack from a deck of
playing cards?
Solution: The individual probabilities are
King =4/52
Queen =4/52
Jack =4/52
Therefore, the probability of success is
4/52+4/52+4/52
=12/52
=3/13
Illustration What is the probability of rolling a die twice and having a 5 and then a 3 show or
having a 2 and then a 4 show?
Solution: The probability of having a 5 and then a 3 show is
P1=1/6,1/6
=1/36
And the probability of having a 2 and then a 4 show is
1/6,1/6
=1/36
1 When tossing a coin, you have what probability of getting either a head or a tail?
2. A bag contains 12 blue, 3 red, and 4 white marbles. What is the probability of drawing
c. in 2 draws, either a red marble followed by a blue marble or a red marble followed by a red
marble?
3. What is the probability of getting a total of at least 10 points in rolling two dice? (HINT: You
want either a total of 10, 11, or 12.)
There are frequently occasions in business when changes in one factor appear to be related
insome way to movements in one or several other factors. For example, a marketing manager
mayobserve that sales increase when there has been a change in advertising expenditure.
Thetransport manager may notice that as Vans and Lorries cover more miles then the need
formaintenance becomes more frequent.
Certain questions may arise in the mind of the manager or analyst. These may be summarized
asfollows:
a) Are the movements in the same pr in opposite directions?
b) Could changes in one phenomenon or variable be causing or be caused by movements inthe
other variable? (This is an important relationship known as a casual relationship).
c) Could apparently related movements come about purely by chance?
d) Could movements in one factor or variable be as a result of combined movements inother
factors or variables?
e) Could movements in two factors be related, not directly, but through movements in athird
variable hitherto to unnoticed?
f) What is the use of this knowledge anyway?
Very frequently, the manager or analysts interested in prediction of some kind. For example,
thequality control manager may want to know what might be the effect on the number of
faultyparts discovered if the amount of expenditure on inspection were increased. The sales
managermay wish to predict sales levels if advertising were increased by say 20%. Here there is
clearlysome kind of casual mode in the minds of the two managers.
Methodology
Suppose that a manager has sensed that two variables or phenomena are behaving in somerelated
way, how might that manager proceed to investigate the matter further? A possiblemethodology
might be as follows.
a) Observe and note what is happening in a systematic way
b) Form some kind of theory about the observed facts
c) Draw a graph or diagram of what is being observed
d) Measure what is happening
e) Use of results
This methodology is developed throughout this chapter and the various stages are
illustratedusing the problem shown in the example 1.
Example 1
Correlation
When the value of one variable is related to the value of another, they are said to be correlated.
Thus correlation means an inter-relationship or association. For example, there is likely to
become correlation between a person’s height and weight.
a) Variables may be;
b) Perfectly correlated (move in perfect unison)
c) Partly correlated (some inter-relationship but not exact)
d) Uncorrelated (no relationship between their movements)
Movements in one variable may cause movements in the same direction in the other variable.
This is known as positive correlation; an example being height and weight. Alternatively,
movement in one variable could cause change in the opposite direction in the other variables.
This is known as negative correlation. For example, if the price of an item is increased, then
fewer will be bought.
The degree of correlation between two variables can be measured and there are two measures of
correlation, denoted by r and R.a. The Product Moment Coefficient of Correlation, denoted by r.
This provides a measure of the strength of association between two variables; one of the
independent variable. R can range from +1, i.e. perfect positive correlation where the variables
change value in the same directions as each other, to -1 i.e. perfect negative correlation where
ydecreases linearly as x increases.
. The Rank Correlation Coefficient, denoted by R
This provides a measure of the association between two sets of ranked or ordered data. R canalso
vary from +1, perfect positive rank correlation to -1, perfect negative rank correlation.
Whichever type of coefficient is being used it follows that a coefficient of zero or near zero
generally indicates no correlation.
Product moment coefficient of correlation (r)
This coefficient gives an indication of the strength of the linear relationship between two
variables.
R= n∑xy-∑x∑y
n∑x^2-(∑x)^2 xn∑y2-(y)^2
15 60 2253600900
24 45 576 20251080
25 50 625 25001250
30 35 90012251050
35 42 1225 17641470
40 46 160021161840
45 28 202 7841260
65 204225400 1300
70224900484 1540
Solution
R= 10x1840-424x363
10x21926-(424)^2x10x15123-(363)^2
R=-0.93
Thus the correlation coefficient is -0.93 which indicates a strong negative linear association
between expenditure on inspection and defective parts delivered. It will be seen that the formula
automatically produces the correct sign for the coefficient.
Note: a strong correlation between two variables would produce an r value in excess of +0.9 or
-0.9. If the value was less than, say 0.5 there would only be a very weak relationship between the
variables.
Least squares
To find the line of best fit mathematically it is necessary to calculate a line which minimizes the
total of the squared deviations of the actual observations from the calculated line. This is known
as the method of least squares or the least squares method of linear regression.
Least squares regression analysis gives equal importance to all the items in the time series, the
older and the more recent. Consequently if the data in the recent past were obtained from
conditions significantly different from long past conditions then it is unlikely that good forecasts
will be achieved using least squares regression analysis. It is because of this that forecasts based
on regression analysis should only be made for the near to medium term future.
Calculating the values of a and b
In the general form of the equation for a straight line
Y = a + bx
Where value a and b are constants and a represents the fixed element and b the slope of the line
i.e. the ratio of the vertical increase in y to horizontal increase in x.
To find a and b it is necessary to solve two simultaneous equations known as the Normal
Equations which are
an + bΣx= Σy……………..equation 1
aΣx+bΣx2=Σxy…………...equation 2
Where n= number of pairs of figures
Note: the slope of the line, b, is sometimes called the regression coefficient.
The use of these equations will be demonstrated using the Example 1 data contained in Table 1
The equations become
10a + 424b = 363
b=n ∑xy-∑ x ∑y
n∑ x^2 – (∑ x)^2
It is often more convenient to use these alternative forms especially when using a calculator.
Value for a and b are re - calculated using the transposed formulae and the table 1 data
When the values have been calculated for a and b, predictions forecasts can be made for values
of x which have not yet occurred. The predictions can be read from the graph on which the line
of best fit has been plotted or the values inserted into straight line formula.
Reverting to example 1 it will be recalled that the manager wished to know the likely number of
defects if 50 p per 1,000 was spent on inspection. From the figure of line of benefit it will be
seen that the number of defects would be 31 per 1,000. The formula can also be used thus:
y = 63.97- 0.65x
So when x is 50
y = 63.97- 0.65(50) = 31.47
Great care must be taken with any form of prediction otherwise absurd results may be obtained.
While any value of x can be used mechanically to make a prediction this does not necessarily
make a practical forecast. It is important to realize that any prediction, for example the 31.47
calculated above, is no more than a single or point estimate and like the use of sample means as
an estimate of the population mean, can be qualified by the use of a confidence interval.
Note; predictions outside the observed values are known as extrapolations predictions within the
observed range are known as interpolations.
350
=
4
= 87.5
The arithmetic mean is very useful because it represents the values of most observations in the
population.
The mean therefore describes the population quite well in terms of the magnitudes attained by
most of the members of the population
= 2245.72 hours
Required calculate the mean age of the students using the coded method
= 29.36 years
NB. The following statistical terms are commonly used in statistical calculations. They must
therefore be clearly understood.
i) Class limits
These are numerical values which limits uq extended of a given class i.e. all the observations in a
given class are expected to fall within the interval which is bounded by the class limits e.g. 15 &
19 are class limits as in the table of the example above.
2. The mode
- This is one of the measures of central tendency. The mode is defined as a value within a
frequency distribution which has the highest frequency. Sometimes a single value may not
exist as such in which case we may refer to the class with the highest frequency. Such a class
is known as a modal class
- The mode is a very important statistical value in business activities quite often business
firms tend to stock specific items which are heavily on demand e.g. footwear, clothes,
construction materials (beams, wires, iron sheets e.t.c.
- The mode can easily be determined form ungrouped data by arranging the figures given
and determining the one with the highest frequency.
- When determining the values of the mode from the grouped data we may use the
following methods;-
i. The graphical method which involves use of the histogram
ii. The computation method which involves use of formula
Example
In a social survey in which the main purpose was to establish the intelligence quotient (IQ) of
resident in a given area, the following results were obtained as tabulated below:
Required
Calculate the modal value of the IQ’s tabulated above using
i. The graphical method and
ii. Formulae
Graphical method
50
40
30
20
10
= 69.14
3. The median
- This is a statistical value which is normally located at the center of a given set of data which
has been organized in the order of magnitude or size e.g. consider the set 14, 17, 9, 8, 20, 32,
18, 14.5, 13. When the data is ordered it will be 8, 9, 13, 14, 14.5, 17, 18, 20, 32
The middle number/median is 14.5
- The importance of the median lies in the fact that it divides the data into 2 equal halves. The
no. of observations below and above the median are equal.
- In order to determine the value of the median from grouped data. When data is grouped the
median may be determined by using the following methods
Example
Referring to the table in 105, determine the median using the methods above
The graphical method
xv
160
140
120
100
80
60
40
20
20 40 60 80 100 120 140 160
n+1
2 Cfbm
Median = L + ×c
cfmc
Where L = Lower class boundary of the class containing the median
N = No of observations
Cfo= cfbm = Cumulative frequency of the class before that containing
the median
F1 fmc = Frequency of the class containing the median
n+1
- Cfbm
Therefore median = L + 2
cfmc
73.5 - 56
= 60 + × 20
48
= 60 + 7.29
= 67.29
4. Geometric mean
- This is a measure of central tendency normally used to measure industrial growth rates.
- It is defined as the nth root of the product of ‘n’ observations or values
i.e. GM = n x1 × x 2 ×... × x n
-
Example
In 1995 five firms registered the following economic growth rates; 26%. 32% 41% 18% and
36%.
Required
Calculate the GM for the above values
GM = 5 26× 32× 41× 18× 36
5
= 1 Log26 + Log32 + Log41 + Log18 + Log26
No. Log
26 1.4150
32 1.5052
41 1.6128
18 1.2533
36 1.5563
Prepared by Mr. Kiarie Page 53
Frontier institute of professional & mangt’studies -Garissa
7.3446
Therefore Log of GM = 1/5 x 7.3446 = 1.46892
So GM = Antilog of 1.46892
= 29.43
5. Harmonic mean
This is a measure of central tendency which is used to determine the average growth rates for
natural economies. It is defined as the reciprocal of the average of the reciprocals of all the
values given by HM.
1
HM =
1 ( 1 + 1 +..... 1
n x1 x2 xn
Example
The economic growth rates of five countries were given as 20%, 15%, 25%, 18% and 5%
Calculate the harmonic mean
1
The HM =
1 (1 + 1 + 1 + 1 + 1
5 20 15 25 10 5
1
=
0.2(0.05+ 0.07+ 0.04 + 0.10+ 0.2)
1
=
0.092
10.86%
6. Weighted mean
- This is the mean which uses arbitrarily given weights
- It is a useful measure especially where assessment is being done yet the conditions prevailing
are not the same. This is particularly true when assessment of students is being done given
that the subjects being taken have different levels of difficulties.
Examples
The following table shows that marks scored by a student doing section 3 and 4 of CPA
LAW 80 35 2800
QT 69 55 3795
FA3 55 60 3300
w = 285 wx = 18005
Weighted mean
Ewx
Ew
18005
285
63.17%
The mode
Merits
i. It can be determined from incomplete data provided the observations with the highest
frequency are already known
ii. The mode has several applications in business
iii. The mode can be easily defined
iv. It can be determined easily from a graph
Demerits
i. If the data is quite large and ungrouped, determination of the mode can be quite
cumbersome
ii. Use of the formula to calculate the mode is unfamiliar to most business people
iii. The mode may sometimes be non existentsor there may be 2 mode for a given set of
data. In such a case therefore a single mode may not exist
The median
Merits
i. It shows the centreof a given set of data
ii. Knowledge of the determination of the median may be extended to determine the
quantities
iii. The median can easily be defined
iv. It can be obtained easily from the cumulative frequency curve
v. It can be used in determining the degrees of skew ness see later
Demerits
i. In some situations where the no. of observations is even, the value of the median
obtained is usually imaginary
ii. The computation of the median using the formular is not well understood by most
business men
iii. In business environment the median has got very few applications
Demerits
i. The determination of the GM by using logarithms is not familiar process to all those
expected to use it manager
ii. If the data contains zeros or –vevalues, the GM ceases to exist
Measures of Dispersion
- The measures of dispersion are very useful in statistical work because the indicate
whether the rest of the data are scattered around the mean or away from the mean.
- If the data is approximately dispersed around the mean then the measure of dispersion
obtained will be small therefore indicating that the mean is a good representative of the
sample data. But on the other hand, if the figures are not closely located to the mean then
the measures of dispersion obtained will be relatively big indicating that the mean does
not represent the data sufficiently
- The commonly used measures of dispersion are
a) The range
b) The absolute mean deviation
c) The standard deviation
d) The semi – interquartile and quartile deviation
e) The 10th and 90th percentile range
f) Variance
a) The range
- The range is defined as the difference between the highest and the smallest values in a
frequency distribution. This measure is not very efficient because it utilizes only 2 values
in a given frequency distribution. However the smaller the value of the range, the less
dispersed the observations are from the arithmetic mean and vice versa
- The range is not commonly used in business management because 2 sets of data may
yield the same range but end up having different interpretations regarding the degree of
dispersion
Example 1
In a given exam the scores for 10 students were as follows
Student Mark (x) x-x
A 60 1.8
B 45 16.8
C 75 13.2
D 70 8.2
E 65 3.2
F 40 21.8
G 69 7.2
H 64 2.2
I 50 11.8
J 80 18.2
618 104.4
Required
Determine the absolute mean deviation
618
Mean, x’ = 10 = 61.8
/ X - X/ 104.4
= = 10.44
Therefore AMD = N 10
Example 2
The following data was obtained from a given financial institution. The data refers to the loans
given out in 1996 to several firms
Required
Calculate the mean deviation for the amount of items given
fx 459,000
X' = = = 24157.9
f 19
/ X - X'/ 286736.90
AMD = =
f 19
= Shs 15,091.40
NB if the absolute mean deviation is relatively small it implies that the data is more compact and
therefore the arithmetic mean is a fair sample representative.
iii. The standard deviation reflects an accurate impression of how much the sample data
varies from the mean. This is because its suitability can also be tasted using other
statistical methods
Example
A sample comprises of the following observations; 14, 18, 17, 16, 25, 31
Determine the standard deviation of this sample
Observation.
x [x-x’)2
14 37.21
18 4.41
17 9.61
16 16.81
25 24.01
31 118.81
121 210.56
121
X' = = 20.1
6
∑ ( x −x ' ) 2 = 286736. 90
∴ standard deviation,
σ=
√ n 6
= 5.93
Alternative method
x X2
14 196
18 324
17 289
16 256
25 625
31 961
121
Example 2
The following table shows the part-time rate per hour of a given no. of laborers in the month of
June 1997.
Calculate the standard deviation from the above table showing how the hourly payment were
varying from the respective mean
fx - fx
2 2
∴standard deviation, σ = f f
2345300 8410
2
-
= 35 35
= 67008.6 577372
= 9271.4
= 96.29
Example 3.1
The quality controller in a given firm had an accurate record of all the iron bars produced in may
1997. The following data shows those records
fx - fx
2
2
∴standard deviation, σ = f f
47489526 118981.50
-
= 313 313
= 84.99 cm
fd
2
fd 2
∴ Standard deviation, σ = f f
2267500 1450
2
= 313 313
= 7244.40 21.50
= 7222.90
= 84.99 cm
C = 50
Standard deviation using the coded method. This is the most preferable method among the three
methods
fu - fu
2
2
σ =c× f f
907 29
2
= 50 × 313 313
= 50 × 1.6997
= 84.99
Variance
Square of the standard deviation is called variance.
Q3 - Q1
SIR =
2
Example 1
The weights of 15 parcels recorded at the GPO were as follows:
16.2, 17, 20, 25(Q1) 29, 32.2, 35.8, 36.8(Q2) 40, 41, 42, 44(Q3) 49, 52, 55 (in kgs)
Required
Determine the semi interquartile range for the above data
Q3 Q1 44 - 25 19
SIR = = = = 8.5
2 2 2
Required
i. Determine the semi interquartile range for the above data
ii. Determine the minimum value for the top ten per cent.(10%)
iii. Determine the maximum value for the lower 40% of the retirees
Solution
The lower quartile (Q1) lies on position
N+1 382 + 1
=
4 4
= 95.75
(95.75 - 50)
the value of Q1 = 29.5 + x 10
69
= 29.5 + 6.63
= 36.13
= 61.08
Q3 - Q1
The semi interquartile range = 2
61.08 - 36.13
=
2
= 12.475
= £12,475
ii. The top 10% is equivalent to the lower 90% of the retirees
The position corresponding to the lower 90%
90
= (n + 1) = 0.9 (382 + 1)
100
= 0.9 x 383
= 344.7
∴the benefits (value) corresponding to the minimum value for top 10%
344.7 - 331
= 69.5 + 40 x 10
= 72.925
= £ 72925
= 153.20
153.2 -119
= 39.5 + 70 x 10
= 39.5 + 4.88
= 44.38
= £ 44380
Example
Using the above data for retirees calculate the 10 th = 90th percentile. The tenth percentile 10th
percentile lies on position
10
100 (382 + 1) = 0.1 x 383
= 38.3
(38.3 x 10)
= 19.5 +
50
= 19.5 + 7.66
= 27.16
The 90th percentile lies on position
90
(382 + 1) = 0.9 x 383
100
= 344.7
∴the value corresponding to the 90th percentile
344.7 - 331
= 69.5 + 40 x 10
= 69.5 + 3.425
= 72.925
∴The required value of the 10th – 90th percentile = 72.925 – 27.16 = 45.765
Revision Exercise
16 0.21
17 0.15
Total 1
(a) If a student is randomly selected from this school, find the probability that
"Elaine gets quiz grades of 79, 82, and 69. She gets a 65 on her final exam. Find the weighted
mean if the quizzes each count for 10% and the final exam counts for 70% of the final grade.
5. Average cost of 5 pencils and 4 rubbers is Rs. 36. The average cost of 7 apples and 8 mangoes
is Rs. 48. What is the total cost of 24 apples and 24 mangoes?
Calculate G.M. and H.M. for the 0—10 10—20 20—30 30—40 40—50
following data : Value :
Frequency : 8 12 20 6 4
(Ans.: 49)
7. A person walks 8 km. at 4 km. an hour, 6 km. at 3 km. an hour and 2 kms. at 4 km. an hour.
Find his average speed per hour
8. Draw a histogram for the following data and locate the mode:
(b) Draw and label the Cumulative Frequency graph for this data
9. (a) A man travels first 900 kms. Of his journey by train at an average speed of 80 kms. Per
hour, next 2,000 kms. By plane at an average speed of 300 kms. Per hour and finally, 20 kms. By
Taxi at an average speed of 30 kms. Per hour. What is his average speed for the entire journey?
(Ans.: 157.24)