-1-

TRIGONOMETRY

1. UNIT CIRCLE: A unit circle is a circle of radius 1 unit. It is usually drawn on the
Cartesian plane with its centre at the origin (0,0). i.e.

More usually, the unit circle is drawn on a grid, e.g.

An angle in standard position has one ray on the positive X-axis (the initial ray) and
the other (the terminal ray) in any of the quadrants depending on the size and direction
of the angle, e.g.

NOTE: Anti-clockwise angles « positive and clockwise angles « negative.

To determine the co-ordinates of any point P, for any angle q in standard position – use a
right-angled triangle.
Remember:

,
but Hyp = 1,
\ sin  = Opp
Þ y-co-ordinate
Similarly:

,
\ cos  = Adj
Þ x-co-ordinate
Thus, the co-ordinates of P are (cos , sin ).
 -2-
Hence, the trigonometric functions can be redefined as cosine is the x-co-ordinate of
the point of intersection of the terminal and the unit circle of an angle in standard
position, and sine is the y-co-ordinate of the point of intersection of the terminal and the
unit circle of an angle in standard position. The unit circle can then be used to determine
the sine and/or cosine of an obtuse angle.

If β is an obtuse angle, then sin  = sin (180 – ), and cos  = -cos (180 – ).

E.g.1. Find the appropriate value for the supplement of:

a) sin 95°
b) cos 130°
c) sin 160°

a) sin 95° = sin (180 – 95)

= sin 85

b) cos 130° = -cos (180 – 130)

= -cos 50

c) sin 160° = sin (180 – 160)

= sin 20

Ref: Ex.1A Q.1-18 (even)

2. AREA OF A TRIANGLE: Consider this triangle. Suppose we know the lengths of a

and c and the size of ÐB, the area can be found without knowing the height.
Draw AD ^ BC.
So, in DADB

Þ h = c ´ sin B
But the area of a triangle is A(D) = ½ ´ base ´ height
By substituting h = c ´ sin B
So, A(D) = ½ ´ a ´ (c ´ sin B)
Þ A(D) = ½ ´ ac ´ sin B
Likewise A(D) = ½ ´ ab ´ sin C or A(D) = ½ ´ bc ´ sin A.
 -3-
E.g.2. Determine the area of DXYZ.

Area = ½ xy sin Z
= ½  18  15  sin 70
 126.9 m²

E.g.3. Determine the value of the variable in each of these triangles given:
a) the area is 15 cm2,
b) the area is 7 m2.

a) ½  a  4  sin 35 = 15

a  13.1 cm

b) ½  3  5  sin  = 7

  69

NOTE: When using sin-1 there is always the possibility of a double solution.

However, if all three side lengths are known but no included angle is known, then

, where . This is known as Heron’s

(or Hero’s) formula.

Ref: Ex.1B Q.1-15 (odd)

3. SINE RULE: Not all triangles are right-angled, so, we need a method to determine the
length of an unknown side or the size of an unknown angle that works with non-right-
angled triangles. One method is the Sine Rule.

The Sine Rule determines the length of an unknown side given two angles and a
corresponding side length, or the size of an unknown angle given the length of two
sides and a corresponding angle.
 -4-
Consider this triangle.
Suppose we know sÐA, sÐC and a,

Þ h = [Link] A

&
Þ h = [Link] C
\ [Link] A = [Link] C

Þ or

Thus the Sine Rule states that for any non-right triangle ABC,

or .

Always check for two solutions when finding an angle size using the Sine Rule.

E.g.4. Solve for ÐN, in DKMN, if n = 5 cm, m = 4 cm, sÐM = 30°.

sN  38.7 or 141.3

In the previous example, there were two possible sizes for angle Y as two triangles can
be drawn that satisfy the given conditions.
 -5-
4. COSINE RULE: Another method that works with non-right-angled triangles is the
Cosine Rule. The Cosine Rule determines the length of the third side given the lengths
of the other two sides and the included angle, or the size of an angle given the lengths of
the three sides.

Consider this triangle.

Given sÐC, a and b,
c2 = h2 + d2 & d = b – e
Þ d2 = (b – e)2
= b2 – 2be + e2
e2 = a2 – h2

&
Þ e = [Link] C
\ c2 = h2 + [b2 – [Link] C + a2 – h2]
= h2 + b2 – [Link] C + a2 – h2
\ c2 = a2 + b2 – 2ab cos C

Thus, the Cosine Rule states that for any non-right triangle ABC, c2 = a2 + b2 – 2ab cos C,

or .

NOTE: With Cosine Rule there is only one solution for an angle.

E.g.5. Given x = 14, y = 11, z = 20, find the size of the smallest angle of DXYZ.

 0.8482
Y  31.98
s

Ref: Ex.1C Q.1-42 (even)

 -6-
INDICES

1. INDIEX LAWS: For am, m is called the index, exponent or power, and a is the base.
The Index Laws/Rules, for a  0, are –

 am  an = am+n

 am  an = am-n

 a0 = 1

 (am)n = amn

 (ab)n = anbn

E.g.1. Without a calculator, evaluate:

a) 810.25

b)
c) (58 ÷ 55)⅓

d)

a) 810.25 =
=3
 -7-

b) =
= 34
= 81

c) (58 ÷ 55)⅓ = (53)⅓

=5

d) =
=  25
=  32

E.g.2. Simplify, expressing your answers in terms of positive indices:

a) c-3d4 ´ c7d-8

b)
c) (27m-6n12)⅓

d)

a) c-3d4 ´ c7d-8 = c4d-4

b) = (p3q-3)-6
= p-18q18

c) (27m-6n12)⅓ = 3m-2n4

d)

Ref: Ex.2A Q.1-124 (R.H.S.); 125-133

 -8-
2. SOLVING EQUATIONS INVOLVING INDICES: Equations involving indices could
have the unknown as the index or as the base. They can be solved by expressing both
sides of the equation as powers of the same base; by reducing the power to 1 for the
unknown; or by refining the solution by a “trial and adjustment” method.

E.g.3. Solve:
a) 8x-1 = 163x
b) a¾ = 27
c) 5x = 100
d) 32x – 7 = 74

e)

a) 8x-1 = 163x
(2³)x-1 = (24)3x
23x-3 = 212x
3x – 3 = 12x
9x = -3
x = -⅓

b) a¾ = 27
(a¾) = (27)

a = 34
a = 81

c) 5x = 100
But 52 = 25 and 53 = 125
Hence, x must be estimated.
Try: x = 2.5, 52.5  56
x = 2.8, 52.8  91
x = 2.9, 52.9  106
 x must be between 2.8 and 2.9, so try:
x = 2.85, 52.85  98
x = 2.86, 52.86  100
 x  2.86

d) 32x – 7 = 74
32x = 81
32x = 34
2x = 4
x=2
 -9-

e)
x⅔ = 4

x = 23
x=8

Ref: Ex.2B Q.1-54 (R.H.S.); 55

 - 10 -
SETS, PROBABILITY AND TREE DIAGRAMS

1. DEFINITIONS AND NOTATION FOR SETS: A set is a group of things that have a
relationship between them. Sets are usually contained between curly brackets { } and
are named using a capital letter. Sets may be represented by listing the elements
between curly brackets, in words, graphically/diagrammatically or by using set
builder notation, e.g. P = {x: x > 2}. A letter with a subscript, e.g. x3, often denotes an
element of a set.

Some symbols associated with sets are –

 is an element of
 is not an element of
 is a subset of
 is a proper subset of
 is not a (proper) subset of
U the universal set, i.e. the set of all things under discussion
… and so on
n(A) or | A | the number of elements in the set A
A' orA the complement of the set A
{ } or  the empty set
A = B the set A is equal to the set B, i.e. the sets contain the same elements

NOTE: Any set which ends with …, e.g. {1, 2, 3,…} is called an infinite set.

A Venn Diagram is a diagrammatic representation of sets.

This diagram shows two sets –
A the even integers from U and
B the prime numbers in U
where U = {2, 3, 4, 5, 6, 7, 8, 9}.

Venn diagrams can be used to illustrate operations between sets –

AÇB

AÈB

A' or`A

Ref: Ex.3A Q.1; 2-16 (even)

 - 11 -
2. VENN DIAGRAMS: Venn diagrams are generally used to solve problems when there
is no particular interest in specific elements, just the number of elements, or the
probabilities for each section of the diagram is important.

E.g.1. This diagram shows the methods of transportation to work of 56 people

interviewed. How many of those interviewed had:
a) only used one method of transport?
b) taken the bus but not used a car?
c) used none of these methods of transport?

a) n(one method) = 10 + 18 = 28

b) n(bus but not car) = 10

c) n(no method) = 8

E.g.2. 100 people were surveyed about which brand of chocolate milk they prefer. The
probabilities are given in this Venn diagram. What is the probability of a person:
a) preferring Brand A?
b) liking neither?
c) liking both equally?

a) P(Brand A) = 0.5

b) P(neither) = 0.25

c) P(both) = 0.15

Ref: Ex.3B Q.1-12 (even)

3. TREE DIAGRAMS: Tree diagrams can show the probabilities associated with the
given sets, or be used to determine probabilities or have the probabilities listed rather
than the numbers of each section.

E.g.3. A bowl contains 6 pink and 7 orange balls. Two balls are selected at random
from the bowl. Determine the probability of obtaining:
a) two pinks?
b) a pink then an orange?
c) a pink given the first was an orange?

a) P(two pinks) =
 - 12 -

b) P(pink then orange) =

c) P(pink| orange 1st) =

= ½

Ref: Ex.3C Q.1-6 (even)

4. COMPOUND EVENTS: When two events are combined, the result is known as a
compound event. The key words are AND and OR. A and B – both events are
satisfied, A or B – one or both events are satisfied. Thus, and is the intersection of the
two sets, and or is the union of the two sets.

E.g.4. When a red and blue dice are rolled simultaneously, what is the probability of:
a) a 4 on the blue die and a 2 on the red die?
b) a 2 on the red die or a sum of 4?

a) P(B4 and R2) =

b) P(R2 or sum of 4) =

Ref: Ex.3D Q.1-6 (even)

ACTIVITIES p.71
 - 13 -
COUNTING

1. ARRANGEMENTS: Counting techniques are used to examine the different outcomes

of choice problems. The multiplication principle determines the number of different
ways (arrangements) of performing one operation followed by another operation. If two
successive events can be performed in “a” ways and in “b” ways respectively, then there
are a  b ways of performing the events in succession.

E.g.1. A librarian has 6 different books to arrange on a shelf. How many different
arrangements are possible?

6 5 4 3 2 1

 n(arrangements) = 6  5  4  3  2  1 = 720

E.g.2. Using the digits 1, 2, 3, 4, 5 and 6 and the letters A, B, C, D and E, how many
six-digit number plates can be made using 3 letters followed by 3 digits?

5 5 5 6 6 6

 n(arrangements) = 53  63 = 27 000

Ref: Ex.4A Q.1-15 (odd)

2. ADDITION PRINCIPLE: If either of two separate (mutually exclusive) events can be

performed in “a” ways and “b” ways respectively, then there are a + b ways of
performing either event, i.e. of performing one event or the other event, provided events
A and B are mutually exclusive. Mutually exclusive events cannot occur together.

E.g.3. Using the digits 0-7 and the letters A-J, how many six-digit number plates can be
made using 6 letters or 6 digits if no letter or digit can be used more than once?

Letters: 10 9 8 7 6 5

Digits: 8 7 6 5 4 3

 n(arrangements) = 10  9  8  7  6  5 + 8  7  6  5  4  3 = 171 360

Ref: Ex.4B Q.1-13 (odd); 14, 15

 - 14 -
E.g.4. How many arrangements of the letters in the word MATHS are possible if:
a) all letters are used?
b) all letters are used but the last must be S?
c) three letters are used?
d) five letters are used but the first must be M and the last must be S?
e) three letters are used but S is excluded?
f) all letters are used but M and S must be together?

a) 5  4  3  2  1 = 120

b) 4  3  2  1 = 24

c) 5  4  3 = 60

d) 1  3  2  1  1 = 6

e) 4  3  2 = 24

f) M and S are treated as one letter and there are two ways of ordering them – MS or SM.
 (4  3  2  1)  2 = 48

Ref: Ex.4C Q.1-11 (odd)

E.g.5. Using the digits 1, 2, 3, 4, 5 and 6:

a) How many six-digit even numbers can be made?
b) How many of the numbers from a) are bigger than 400 000?

a) For an even number the last digit must be a 2, 4 or 6.

? ? ? ? ? 3

This leaves 5 digits to choose from for the first, 4 for the second, …
5 4 3 2 1 3

Thus the number of six-digit even numbers = 5  4  3  2  1  3

= 360

b) For numbers to be greater than 400 000, the first digit must be a 4, 5 or 6.
Numbers > 400 000 and ending in a 2: 3 4 3 2 1 1

Numbers > 400 000 and ending in a 4: 2 4 3 2 1 1

Numbers > 400 000 and ending in a 6: 2 4 3 2 1 1

 The total number of even numbers > 400 000 = 72 + 48 + 48
= 168

Ref: Ex.4D Q.1-10 (even)

 - 15 -
3. PROBABILITY: Counting techniques are sometimes useful in probability problems.

E.g.6. If two letters are picked from the alphabet, what is the probability that the two
letters are A and T?

Total number of 2 letter picks = (26  25)  2

= 325

E.g.7. If all the letters of the word MATHS are arranged in random order, what is the
probability that:
a) the M will be the first letter?
b) the T and H will be next to one another?
c) the A will be first or last?
d) the S will be last given the M was first?

Total number of arrangements = 5  4  3  2  1

= 120

a) P(M1) =

b) P(TH or HT) =

c) P(A1 or A5) =

d) P(S5 | M1) =

=¼

Ref: Ex.4E Q.1-13 (odd)

 - 16 -
FUNCTION

1. SKETCHING PARABOLAS: The graph of a quadratic function is a parabola. The

two most important features of a parabola are its axis (line) of symmetry and its
turning point (maximum or minimum) which lies on the axis of symmetry. A
“standard” quadratic equation is y = x2, where the turning point is (0,0), and it has a
minimum turning point.

For an equation in the form of y = a[b(x – c)2] + d –

+a dilates a “standard” graph parallel to the Y-axis,
-a reflects a “standard” graph in the X-axis,

b dilates a “standard” graph parallel to the X-axis with a scale factor of ,

c translates a “standard” graph horizontally c units to the left for +c and the right for -c,
d translates a “standard” graph vertically d units up for +d and down for -d.

E.g.1. Sketch the graph of:

a) y = (x – 1)2 + 3
b) y = 2(x + 3)2 – 1
c) y = -⅓(x – 4)2 – 2

a) y = (x – 1)2 + 3
Translate graph right 1 unit and up 3 units.

b) y = 2(x + 3)2 – 1
Dilate graph parallel to the Y-axis, and
translate graph left 3 units and down 1 unit.

c) y = -⅓(x – 4)2 – 2
Dilate graph parallel to the Y-axis, reflect
graph in the X-axis and translate graph right
4 units and down 2 units.

Ref: Ex.5A Q.1-7

2. OTHER FUNCTIONS: The general form of a cubic function is given by

ax3 + bx2 + cx + d where a, b, c and d are constants and a ¹ 0. The graph of a cubic
polynomial is always a continuous curve and may cut the X-axis once, twice or three
times. For +a the graph is in the 1st and 3rd quadrants, and for –a the graph is in the 2nd
and 4th quadrants.

NOTE: If a function is given in factorized form then the roots of the function are
determined by equating each term to zero.
 - 17 -

The general form of a root function is given by where if n is even, then the
graph is in the 1st quadrant, but to get a “complete” graph then graph ,
which is equivalent to a “parabola” in the X-axis, and if n is odd, then the graph is in
the 1st and 3rd quadrant, which is equivalent to a “cubic” in the X-axis.

The general form of a reciprocal function is given by or xy = k. For +k the

graph is in the 1 and 3 quadrants, and for –k the graph is in the 2nd and 4th
st rd

quadrants. The graph is asymptotic to both the X- and Y-axes.

The general form of an exponential function is given by y = [Link] where k is the value of
the vertical intercept. The graph is asymptotic to the X-axis and is only in the 1st and
2nd quadrants, with the graph is ‘high’ in the 1st quadrant for +k, and the graph is
‘high’ in the 2nd quadrant for –k. All of these functions conform with the
transformations studied for quadratic
2– 10
4
6
8
10
4
6
8
10
2
4
6
8
2– 10
2
4
6
8 functions.

E.g.2. Sketch the graph of:

a) f(x) = (x + 2)(x – 1)(x – 4)
y y = 3 x + 2
b) 10
8
6
c) 5 4
y = + 2
x 2
a) f(x) = (x + 2)(x – 1)(x – 4) x
– 10 – 8 – 6 – 4 – 2 2 4 6 8 10
–2
–4 f(x) = (x + 2)(x – 1)(x – 4)
b) –6
–8
– 10

c)

Ref: INVESTIGATION p.107

Ex.5B Q.1; 2-12 (even)

3. TRANSFORMATIONS: Some types of transformations are –

 a translation in the Y direction, e.g. y = f(x)  y = f(x)  c,
 a translation in the X direction, e.g. y = f(x)  y = f(x  b),
 a reflection in the Y-axis, e.g. y = f(x)  y = f(-x),
 a reflection in the X-axis, e.g. y = f(x)  y = -f(x),
 a dilation in the Y direction, e.g. y = f(x)  y = a.f(x), and

 a dilation in the X direction, e.g. y = f(x) 

Graphs are often a combination of more than one transformation.

 - 18 -
E.g.1. The graph on the right shows y = f(x). Draw the graph of:
a) y = -f(x) – 1
b) y = 2f(-x)
c) y = f(x + 3)
d) y = f(2x + 2)
e) y = f(1 – x)

a) y = -f(x) – 1

b) y = 2f(-x)

c) y = f(x + 3)

d) y = f(2x + 2)
i.e. 2x + 2 = 4  x = 1
For more
2x + 2 = 3  x = complex
2x + 2 = 0  x = -1 transformations,
this method
e) y = f(1 – x) transforms the
1 – x = 4  x = -3 significant
1 – x = 3  x = -2 points of
1–x=0  x=1 interest.

Ref: TRANSFORMATION GRAPHS

4. NOTEWORTHY FEATURES: Consider this graph:

Point A is the global minimum. Points B, D and H are roots.
Point C is a (local) maximum point. Points E and I are
points of inflection. Point F is the y-intercept. Point G is a
(local) minimum point. Point J is a point of horizontal
inflection. Point K is the global maximum.

The terms concave down and concave up are often used to describe the shape of a graph
near a maxima or a minima. Points on the curve where it changes from being concave
down to concave up, or vice versa, are called points of inflection.
 - 19 -
The domain is the set of x-numbers that a function can use, and the range is the set of
y-numbers that result from the domain. If no specific domain is given, then assume it to
be all of the Real numbers. For a variety of reasons, a restricted domain is sometimes
given for a particular function.

E.g.4. a) Find the range of the function f(x) = 2x2 – 5 given the domain is {-1, 0, 1, 2}.
b) State the domain of the function g(x) = given the range is {-1, 0, 1, 2}.

a) f(x) = 2x2 – 5  Domain = {-1, 0, 1, 2}

f(-1) = -3 f(0) = -5 f(1) = -3 f(2) = 3
 Range = {-3, -5, 3}

b) g(x) =  Range = {-1, 0, 1, 2}

= -1  x = -½ =0x=0 =1x=½ =2x=4
 Domain = {-½, 0, ½, 4}

Ref: Ex.5C Q.1-5; 6-10 (even)

5. SOLVING EQUATIONS: Equations can be solved by ‘guess and check’,

algebraically, using any of the solver programs on a calculator or graphically.

To solve graphically – graph all of the equations, and read the appropriate point(s) of
intersection from the graph.

E.g.5. Solve graphically:

a) 2x2 – 6x + 1 = 3x – 4
10
2
4
6
8– 10
2
4
6
8
2– 10
4
6
8
10 2
4
6
8
b)
c)

a) 2x2 – 6x + 1 = 3x – 4 y
2
y = 2x – 6x + 1
From the graph (CAS calc.), 10

x  0.63 8

4 y = x + 5
b) 2
From the graph (CAS calc.), x
– 10 – 8 – 6 – 4 – 2 2 4 6 8 10
x  1.72 x
– 2
y = 3 – 4
– 4

– 6
c) – 8
From the graph (CAS calc.), – 10
x  -0.19 or x  3.29

Ref: Ex.5D Q.1-4

 - 20 -
6. PROPORTION: Two quantities, a and b, are directly proportional if one quantity
increases as the other increases and vice versa. This is written as a  b or a = kb, where
k is the constant of variation. The graph of a direct proportion relationship is a line
passing through the origin (0,0), provided c = 0 with a gradient of k. When graphing a
versus b or a against b – b is on the horizontal axis, and a is on the vertical axis.
Likewise, in a table of values – b is in the top row or left column, and a is in the
bottom row or right column, with a constant first difference pattern.

Two quantities are inversely proportional if one quantity increases as the other

decreases. This is written as a  or a = , where k is the constant of variation. The

graph of an inversely proportion relationship is a curve that is asymptotic to both the
X- and Y-axes, passing through the points (1,k) and (k,1). In a table of values – there is
a constant product, with a  b = k. Inverse proportion is sometimes known as the
“sharing” function. Other relationships are neither direct nor inverse.

E.g.6. “River Extra Energy”, a company which provides mineral water, charges a fixed
delivery fee of $30 plus $1.30 per litre for its product.
a) Write an equation relating C, the charge in dollars, with n, the number of litres
of mineral water delivered.
Another mineral water company, “El Dale” delivers water but without any
delivery charge. The equation which they use to calculate the cost, F dollars, for
n, the number of litres of mineral water delivered, is F = 2n.
b) For what values of n is it better to buy from “El Dale” than from “River Extra
Energy”?

a) C = $(1.3n + 30)

b) From the graph,

It is better to buy from “El Dale” for n  42 L.

Ref: Ex.5E Q.1-9 (odd)

 - 21 -
RATES AND VARIATION

1. RATES: A rate is the amount one quantity varies in relation to another. The two
quantities are different and have different units. Each unit cannot be changed to the
other.

E.g.1. A marathon runner covers the 42 km distance at an average speed of 12 km/h.

a) How long should it take her to finish the race?
What is her speed in:
b) metres per hour?
c) metres per minute? and
d) metres per second?

= 3½ h
 It should take the runner 3½ hours to finish the race.

b) 12 km/h = 12 000 m/h

c) 12 000 m/h = 200 m/min

d) 200 m/min = 3⅓ m/s

Ref: RATES Q.1-22 (even)

2. PROPORTION: A proportion is a comparison between two different quantities. A

proportion exists when two ratios are equal. Two quantities are directly proportional
if one quantity increases as the other increases and vice versa. Two quantities are
inversely proportional if one quantity increases as the other decreases.

The unitary method is one method used to solve proportional problems. The method
involves calculating the value of one unit measure of the ‘known’ quantity, and then
calculating the proportional value of the other quantity.

E.g.2. a) If 10 litres of petrol cost $14.20. What would 6½ litres cost?

b) If 6 painters can paint an office building in 8 days. How long would it take 9
painters working at the same rate?

a) 10 litres of petrol cost $14.20

1 litre of petrol cost $1.42
6½ litres of petrol cost $9.23
 - 22 -
b) 6 painters take 8 days
1 painter takes 48 days
9 painters take 5⅓ days

Ref: UNITARY METHOD Q.1-30 (even)

3. DIRECT VARIATION: If two quantities are directly proportional we write y µ x

which is read as “y is directly proportional to x”. So, if y µ x, then y = kx where k is a
constant called the constant of proportionality.

When two directly proportional quantities are graphed, the result is a line with a
slope of k and which passes through the origin, (0,0), i.e. the y-intercept is zero. As one
variable increases, the other variable increases, and as one variable decreases, the
other variable decreases.

E.g.3. Fruit buns cost 60 cents each.

a) Complete this table.

NUMBER BOUGHT (x) 0 1 2 3 4 5

PRICE PAID (y)

b) Graph this information.

c) Determine the law connecting x and y.
d) State the value of k.
e) Confirm that the quantities are directly proportional.

a)
x 0 1 2 3 4 5
y 0.00 0.60 1.20 1.80 2.40 3.00

b)

c) y = 0.6x
d) k = 0.6
 - 23 -
e) 1 fruit bun costs $0.60
2 fruit buns cost $1.20
i.e. Double the number of buns doubles the cost.
 the quantities are directly proportional.

Ref: DIRECT PROPORTION

4. CONSTANT OF VARIATION: For any point in a relationship that is directly

proportion, the ratio is the constant of proportionality or variation.

E.g.4. If x and y are directly proportional and y = 24 when x = 3, find:

a) the constant of proportion, and
b) the value of x when y = 16.

a) k = 8  y = 8x

b) When y = 16, x = 2

Ref: CONSTANT OF PROPORTIONALITY

5. PARTIAL VARIATION: For x and y, if y varies partly as x and partly as a constant,

then y = kx + c is called the partial proportion or variation. k is still called the
constant of proportionality or variation. Two points are required to find the values of
k and c.

E.g.5. Marina discovers water leaking from a pipe in her kitchen. The plumber’s charge
varies partly with the time (t) it takes him to fix the problem and a call-out fee of
$30. If a 45 minute job costs $90, find:
a) the rule relating cost with time,
b) the cost of a 1 hour and 20 minute job, and
c) how long it took to do a job costing $230?

a) c = $30
Cost = kt + 30
90 = 45k + 30
45k = 60
k = 1⅓
 Cost = 1⅓t + 30

b) 1 hour 20 min = 80 minutes

Cost = 1⅓  80 + 30
 $136.67
 - 24 -
c) 230 = 1⅓t + 30
1⅓t = 200
t = 150 min
 A $230 job takes 2½ hours.

Ref: PARTIAL PROPORTION

6. INVERSE VARIATION: If one quantity increases as another quantity decreases,

then the two quantities are inversely proportional. If two quantities are inversely

proportional, then and or xy = k where k is the constant of

proportionality or variation. The graph of an inverse proportion is an hyperbola, or
the ‘positive half’, where neither variable is equal to zero.

E.g.6. Let x be the number of painters and y the number of days taken to complete a
job.
a) Complete the following table:

x 1 2 3 4 5
y 24

b) Draw a graph of these results.

c) Determine the law connecting x and y.
d) State the value of k.
e) How many painters would be required to complete the job in 3 days?

a)
x 1 2 3 4 5
y 24 12 8 6 4.8

a)

b) xy = 24 or

c)  k = 24
 - 25 -

d)
=8
 It takes 8 painters 3 days to complete the job.

Ref: INVERSE PROPORTION

 - 26 -
CENTRAL TENDENCY AND DISPERSION

1. MEASURES OF CENTRAL TENDENCY: Data is a set of scores. A statistic is a

single number used to represent the data, in some manner. The mean ( or ) is the
sum of the scores divided by the total number of scores. The median of a set of scores is
the middle score, or the mean of the two middle scores, when the data is arranged in
numerical order. The mode is the most frequent score. Not all data have a mode.
Some data have two modes – this is called a bimodal set of data. The mean, median
and mode are all measures of central tendency.

NOTE: The modal value on the graphics calculator is unreliable but the CAS calculator
is very reliable, showing “no mode” and bimodal solutions.

The measures of central tendency can be calculated from the raw data, a frequency
table, or a variety of graphs such as a dot frequency diagram or a stem and leaf plot.

E.g.1. Calculate the measures of central tendency for each of the following:
a)
SCORE FREQUENCY
3 1
4 2
5 1
6 3
TOTAL 7

b)

c)
21 5
22 5 3 8
23 0 1 7 4 2
24 4 4 3 5 1 0
25 8 9 5
26 6 2
 - 27 -
Using the g. calc.
a) Mean  4.86
Mode = 6
Median = 5

b) Mean = 13.2
Mode = 14
Median =13.5

c) Mean = 240.6
Mode = 244
Median 240.5

Ref: Ex.6A Q.1-17 (odd)

2. MEASURES OF DISPERSION: The range is the difference between the highest and
lowest scores. The interquartile range is the difference between the upper and lower
quartiles. Both the range and the interquartile range are some of the measures of
spread (or dispersion).

The deviation of any score is the difference between that score and the mean, i.e. .
The sum of the deviations is always zero. Thus the mean, or more fully the arithmetic
mean, is that value which gives a zero sum for the deviations.

The variance is the average of the squares of the deviations. It is denoted by s2 or 2

and is or . As a measure of dispersion, the variance has

an obvious weakness; it produces results in squared units. This disadvantage can be
overcome by taking the square root of the variance.

The square root of the variance is called the standard deviation. It is the most widely
used statistic because of its mathematical reliability and accuracy. The standard

deviation (s or σ) is the square root of the variance and is or

E.g.2. The scores for the test results of a group of ten students were 3, 5, 5, 5, 6, 7, 7, 8,
8, 9. Find the:
a) range,
b) interquartile range,
c) variance and
 - 28 -
d) standard deviation.
 - 29 -
3 5 5 5 6 7 7 8 8 9
^ ^ ^
L.Q. Med U.Q.

Range = 9 – 3 = 6
Interquartile range = 8 – 5 = 3

Scor Frequency Deviation Deviation²

e f
x
3 1 -3.3 10.89 10.89
5 3 -1.3 1.69 5.07
6 1 -0.3 0.09 0.09
7 2 0.7 0.49 0.98
8 2 1.7 2.89 5.78
9 1 2.7 7.29 7.29
Total 30.10

= 3.01

 1.73

Ref: Ex.6B Q.1-12 (even)

3. DEVIATION INTERVALS: For most distributions, very few, if any, of the scores are
more than three standard deviations from the mean, i.e. the scores lie between
and .

NOTE: Your calculator gives both sn and sn-1. We use sn although sn-1 gives a better
answer as it compensates for there (usually) being less variation in a small
sample than there is in a population.

E.g.3. For the data set 22, 45, 35, 55, 50, 67, 70, 17, 40, 45, 51, 50, 54, 65, find:
a) the mean and standard deviation,
b) how many scores lie less than one standard deviation from the mean, and
c) how many scores lie less than two standard deviations from the mean.
 - 30 -
a) From the g. calc.:x  47.57 and   14.93

b) x –   32.64 andx +   62.50

 9 scores are less than one standard deviation from the mean.

c) x – 2  17.71 andx + 2  77.44

 13 scores are less than two standard deviations from the mean.

Ref: Ex.6C Q.1-15 (odd)

4. GROUPED DATA: For data with a lot of scores but very few repeated values it is often
better to represent the data in groups or class intervals, although some information is
lost. All statistics, except mode, are calculated at the midpoint of the intervals. The
mode is given as the modal class interval. Graphs are drawn on the class boundaries.

Ref: Ex.6D Q.1-11 (odd)

 - 31 -
BOXPLOTS AND HISTOGRAMS

1. BOX AND WHISKER DIAGRAMS: A box plot or a box and whisker diagram is a
summarizing diagram that illustrates both the centre of a distribution (usually the median
although sometimes the mean) and it’s dispersion (both the interquartile range and the
range). The interquartile range (I.Q.R.) takes into account the ranks of all of the scores
and is calculated by I.Q.R. = Q3 – Q1 where Q3 is the upper quartile and Q1 is the lower
quartile. The median is sometimes referred to as Q2.

NOTE: The whiskers show the first and last scores and the box marks the L.Q., median
and the U.Q.

E.g.1. Draw a box plot for the following data: 7, 11, 13, 12, 8, 5, 9, 10, 8, 4, 3, 4, 5, 4, 2.

2, 3, 4, 4, 4, 5, 5, 7, 8, 8, 9, 10, 11, 12, 13

A boxplot can also be used to compare two or more sets of data. Each plot is graphed
above the line and have a gap between the box of each plot. When analysing or
comparing data sets, comment on the location of the median (or mean) scores, the
spread or range/interquartile range of the scores, the “shape” of the distribution in
terms of its symmetry or skewness, outliers, etc.

If the median is in the middle of the box, then the data set is symmetrical. If the median
is towards the right-hand end of the box, then the data set is negatively skewed as more
data is on the left. If the median is towards the left-hand end of the box, then the data set
is positively skewed as more data is on the right.

E.g.2. Compare the data from E.g.1 to the following data:

4, 5, 7, 8, 9, 8, 10, 10, 9, 9, 8, 8, 5, 4, 6.

4, 4, 5, 5, 6, 7, 8, 8, 8, 8, 9, 9, 9, 10, 10

The data in E.g.2 has a smaller dispersion, i.e. a smaller range and interquartile range but
it has a higher median than the data in E.g.1.

Ref: Ex.7A Q.1-10 (even); 5

 - 32 -
2. HISTOGRAMS: A histogram is a “column” graph with adjoining columns. It is used
for grouped discrete data or for continuous data but the data may appear to be
discrete, in nature. If the data appears to be discrete, then the class intervals or “bins”
are from half unit to half unit, e.g. 12.5 ® 13.5, 13.5 ® 14.5, 14.5 ® 15.5, etc., and the
“bin width” is 1 unit. The vertical axis is the frequency and the horizontal axis is a
number line.

The class intervals should be of equal size, and the frequency can, in some cases, be
given as a relative frequency, i.e. a fraction of the whole expressed as a decimal, or as a
percentage frequency, i.e. a percentage of the whole. The relative frequency indicates
how significant a particular score is compared to the total distribution. The percentage
frequency indicates the percentage of the total scores for a given score.

A frequency polygon is a special form of line graph that uses the mid-points of the
groups or intervals. It is often drawn over a histogram. A frequency polygon starts at
the midpoint of the previous interval to the start of the histogram, and ends at the
midpoint of the next interval to the end of the histogram, so the axis must have
sufficient space left for this.

E.g.3. Draw a histogram and frequency polygon for this data.

HEIGHT BOUNDARIES FREQUENCY

1.45-1.49 1.445-1.495 3
1.50-1.54 1.495-1.545 6
1.55-1.59 1.545-1.595 9
1.60-1.64 1.595-1.645 11
1.65-1.69 1.645-1.695 12
1.70-1.74 1.695-1.745 7
1.75-1.79 1.745-1.795 2
TOTAL 50

HEIGHTS
Freq
12
10
8
6
4
2

1.4 1.5 1.6 1.7 1.8

Height (metres)

Ref: Ex.7B Q.1, 4-6

 - 33 -
As for boxplots, histograms can also be skewed. In general, for a negatively skewed
data set – mean < median < mode and for a positively skewed data set –
mode < median < mean.

Ref: Ex.7C Q.1-10 (even); 7

 - 34 -
SEQUENCES

1. SEQUENCES: A sequence is any group or set of numbers, most of which follow some
definite pattern. Each member or term of the sequence can be labelled as T1, T2, T3,...,Tn.
Tn is called the general term and is often used to define the sequence.

Other letters, besides T, are sometimes used particularly for well known sequences. F is
usually used for the Fibonacci sequence. L is usually used for the Lucas sequence.

E.g.1. Find the 7th term of each of the following sequences:

a) Tn = 2n + 5
b) Tn = n² + n – 1

a) Tn = 2n + 5
T7 = 2(7) + 5
= 19

b) Tn = n² + n – 1
T7 = 7² + 7 – 1
= 55

2. RECURSIVE FORMULAE: A recursive formula “runs back” which simply means

that it refers back to previous terms. To determine the sequence – the value of at least
one term, usually the first term, must be known and then any others can be determined
from this, e.g. Tn+1 = Tn + 3 with T1 = 4.

A recursive formula may be re-written to form a difference equation (rule) which

indicates the difference between successive terms, e.g. Tn+1 – Tn = 3 with T1 = 4. A
formula for Tn can usually be determined using a table of values by calculating the
difference pattern and hence, the rule (as was previously done).

E.g.2. Write the difference equation for the sequence 4, 7, 10, 13,… and write the rule
for the nth term.

4, 7, 10, 13, ... difference = 3

 Tn+1 – Tn = 3 and T1 = 4
So, Tn = 3n + 1

3. ARITHMETIC SEQUENCES: An Arithmetic (Sequence or) Progression (A.P.) is a

sequence of terms where there is a common difference (d) between successive terms,
i.e. a linear sequence. In general, the first term of an A.P. is a, so the A.P. is a, a + d,
a + 2d, a + 3d, ... Hence, the nth term will be Tn = a + (n – 1)d.
 - 35 -
E.g.3. Find, for the sequence 21, 18, 15, ...:
a) the common difference,
b) the 7th term, and
c) the nth term.

21, 18, 15,... i.e. a = 21

a) d = 18 – 21 = -3

b) Tn = a + (n – 1)d
T7 = 21 + (7 – 1)(-3) = 3

c) Tn = 21 + (n – 1)(-3)
= 21 – 3n + 3
= 24 – 3n

E.g.4. Find the first 4 terms of the arithmetic sequence in which T5 = 7 and T11 = 22.

T5 = 7 and T11 = 22
 a + 4d = 7 and a + 10d = 22
 a = 7 – 4d a = 22 – 10d
7 – 4d = 22 – 10d
-4d + 10d = 22 – 7
6d = 15
d = 2.5
& a + 4(2.5) = 7
a + 10 = 7
a = -3
 The first 4 terms of the sequence are -3, -0.5, 2, 4.5.

4. GEOMETRIC SEQUENCES: A Geometric (Sequence or) Progression (G.P.) is a

sequence of terms with common ratio (r) between successive terms. In general, the first
term of the G.P. is a, so the G.P. is a, ar, ar², ar3... Hence, the nth term will be Tn = arn – 1.

E.g.5. a) Find the 12th term of the sequence 50, 60, 72, 86.4, ...
b) Which term of the G.P. 2, 6, 18, 54, ... is 39 366?
c) The fifth and eighth terms of a G.P. are 15 and 1 875 respectively. Find the
sequence.

a) 50, 60, 72, 86.4,...

Tn = arn – 1
Tn = 50(1.2)n – 1
T12 = 50(1.2)11
 371.5
 - 36 -
b) 2, 6, 18, 54,...
a=2 r=3
Tn = arn – 1
39 366 = 2(3)n-1
3n – 1 = 19 683

n–1=9
n = 10
 39 366 is the 10th term of the given G.P.

c) T5 = ar4 = 15 and T8 = ar7 = 1 875

r3 = 125
r=5

 The G.P. is 0.024, 0.12, 0.6,...

5. GROWTH AND DECAY: Many number patterns exist in everyday situations and an
understanding of sequences can be useful in solving real life problems.

NOTE: Many money problems involve T0 instead of T1.

E.g.6. An employee is hired on a quarterly salary of $7 000 with increments of $250 per
quarter.
a) What will the employee earn in the tenth quarter of employment?
b) When will the employee be earning $8 500 per quarter?
 - 37 -
a)
QUARTER 1 2 3 4
SALARY 7 000 7 250 7 500 7 750

d 250 250 250

Rule: s = 7 000 + 250(q – 1)

s(10) = 7 000 + 250(9)
= $9 250
 The employee will earn $9 250 per quarter in the tenth quarter of employment.

b) 8 500 = 7 000 + 250(q – 1)

1 500 = 250(q – 1)

6=q–1
q=7
 The employee will be earning $8 500 per quarter in the seventh quarter.

E.g.7. The predicted growth pattern of a town with a current population of 2 000, is for
a population of 2 400 at the end of the first year, 2 880 at the end of the second
year, 3 456 at the end of the third year and so on. Given that the pattern
continues, find:
a) the population at the end of 10 years, and
b) when the population reaches approximately 23 610.

a)
YEAR 0 1 2 3
POPULATION 2 000 2 400 2 880 3 456

d 400 480 576

r 1.2 1.2 1.2

Rule: p = 2 000(1.2)y
p(10) = 2 000(1.2)10
 12 383
 The population will be approximately 12 383 at the end of 10 years.

b) [Graph p = 2 000(1.2)y; then use G-Solv, X-cal]

From the CAS calc., y  13.5 when p = 23 610
 The population will reach approximately 23 610 in the 14th year.
 - 38 -
E.g.8. $1 500 is deposited on 1st January, 1985 to open an account. On each subsequent
anniversary of this date a further $1 500 is deposited into the account. If the
account earns interest at the rate of 10% per annum compounded annually, how
much will be in the account:
a) on 31st December, 1999 immediately after the annual interest calculation, and
b) on 1st January 2000, following the deposit of $1 500?

a) Total account balance (31st December, 1985) = $1 500  1.1

= $1 650

 $52 424.59

b)  Account balance (1st January, 2000)  $52 424.59 + $1 500

 $53 924.59

Ref: Ex.8A Q.1-22 (even)

6. PAYMENTS: Regular payments can be made into an account, e.g. to pay off a loan, or
from an account, e.g. an annuity. Where regular payments are made to pay off a loan –
each payment reduces the amount owing and hence, the interest is reduced. This is
called reducible interest or amortization.

An annuity is a regular payment paid from an account in the form of an allowance or a

pension.

NOTE: A spreadsheet is an efficient method of setting up a repayment schedule.

E.g.9. At the beginning of April, Lindsay borrowed $5 000 to buy a car. She is
considered a medium credit risk and so is required to pay an interest rate of 24%
p.a. calculated monthly. She repays $550 per month with payments due on the
first of each month. In August she receives her tax refund and decides to pay off
the loan (September 1). How much must she pay?
 - 39 -
P = $5 000 Interest rate = 2% per month Payments = $550 per month

AMOUNT INTEREST MONTHLY BALANCE

PAYMENT
OWING ($) PAYABLE ($) PAYMENT ($) OWING ($)
May 1 5 000.00 100.00 550.00 4 550.00
June 1 4 550.00 91.00 550.00 4 091.00
July 1 4 091.00 81.82 550.00 3 622.82
August 1 3 622.82 72.46 550.00 3 145.28
September 1 3 145.28 62.91 3 208.18 ---

 Lindsay must pay $3 208.18 on September 1 to pay off her loan.

Ref: Ex.8B Q.1-5 (odd)

 - 40 -
SAMPLING METHODS AND CAPTURE-RECAPTURE

1. SAMPLING: Some population sizes (N) are too large to survey. So, a small group or
sample is selected to represent the population. As a guide, the sample size (n) is
. There are a number of different sampling techniques. We must ensure
that any sample is truly random, and that each member of the population has an equal
chance of being included in the sample. This is known as simple random sampling. To
choose a true random sample, you could draw names from a hat, roll a die, use
random digit tables, or use the random number generator on your calculator.

When the target population is known to consist of different classes (strata), it is possible
that a small sample may not truly represent the entire population. We would then use a
stratified random sample. The sample size of each stratum is in the same proportion
as the stratum to the population. The advantage with stratified random sampling is
that it takes into account the various chosen strata. A simple random sampling is
conducted on each stratum. It offers a more efficient sampling with the possibility of
smaller sample sizes.

Cluster sampling involves selecting an identifiable group within the population to

represent the population. Surveying a school House group is an example of cluster
sampling.

Judgement sampling involves any sample selected by the surveyor that they believe
accurately reflects the opinion of the population, e.g. A class, House group or Form
group could be used as a sample for a school survey.

Quota sampling involves selecting the required number in the stratified sample by a
non-random method, e.g. Pollsters in Hay Street or a shopping mall, asking ‘random’
people to take part in a survey, but only choosing people that ‘fit’ within the required age
groups and/or sex.

Systematic sampling involves selecting the sample in a predetermined pattern, e.g.

Every 10th person on the school roll.

Convenience sampling involves selecting a sample in the easiest and/or most cost
effective manner, and thus utilizing any of the previous methods, whichever is most
convenient and/or cost effective at the time of the survey.

E.g.1. From a class of 30, use the calculator random digit generator to randomly select 5
students if each student has been allocated a number (1-30).

e.g. From the CAS calc., 3, 2, 25, 22, 30

 - 41 -
E.g.2. Students from a high school are to be surveyed and a sample of 40 is to be
selected. Complete the following table to determine the size of the sample for
each year group.

YEAR 8 9 10 11 12 TOTAL
NUMBER 270 215 190 120 105 900
PROPORTION 0.3 0.238 0.21 0.13 0.116 1
SAMPLE 12 10 8 5 5 40

Ref: Ex.9A Q.1-6 (even)

2. CAPTURE-RECAPTURE: Capture-recapture is a method that makes it possible to

estimate the population of animals, in the wild. The process involves catching and
tagging a sample of the population, releasing the tagged animals back into the
population, and at a later date, capturing another sample, noting the number of tagged
animals in the second sample. Assuming the proportion of tagged animals, from the
first capture, in the population is reflected in the proportion of tagged animals in the
second capture, then an estimate of the total population is

E.g.3. In an attempt to estimate the size of a recently discovered numbat population –

86 were captured, tagged and released. 1 week later, 73 were captured and 8 were
tagged. What is the estimated size of the population?

= 784.75
 The numbat population is estimated to be 785.

Ref: Ex.9B Q.1-4

It is not ‘possible’ to capture and tag people. However, a similar process can be used to

determine human populations where .

 - 42 -
E.g.4. To estimate the number of people requiring knee reconstructions as the result of
sports injuries, a government enquiry compared the number of WAFL/AFL
footballers requiring the reconstruction (1 024) to the number of people receiving
the treatment from Royal Perth Hospital (53) with 34 of the footballers received
their treatment at Royal Perth. Use this information to estimate the total number
of reconstructions during that year.

Let k be the total number of knee reconstructions due to sports injuries.

= 1 596
 Approximately 1 596 knee reconstructions due to sports injuries were in this year.

Ref: Ex.9C Q.1-4

 - 43 -
NORMAL DISTRIBUTION

1. NORMAL DISTRIBUTIONS: Many naturally occurring variables, e.g. human I.Q.’s,

the weights of cats, etc. fit a smooth ‘bell shaped curve’ known as a normal curve.

A normal curve with a mean of zero and a standard deviation of one is known as a
standard normal distribution.

For standard normal distributions –

68.3% of the distribution 95.4% of the distribution 99.7% of the distribution
lies within one standard lies within two standard lies within three standard
deviation of the mean. deviations of the mean. deviations of the mean.

The normal distribution is also called the Gaussian distribution.

To determine the probabilities for normally distributed sets of data, a book of tables or
an appropriate calculator programme may be used. The entries in the book of tables
are cumulative probabilities associated with a particular standardized score z where

. The shaded area to the left of z in this diagram represents the cumulative
probability, where P(Z  z).
 - 44 -
E.g.1. Find:
a) P(Z £ 1.74)
b) P(-0.7 £ Z £ 0.4)
c) P(Z > 0.73)
d) P(Z £ k) = 0.9798
e) P(Z > k) = 0.83
f) P(-1.2 £ Z £ k) = 0.642
g) P(-k £ Z £ k) = 0.68

a) P(Z  1.74) = 0.9591

b) P(-0.7  Z  0.4) = P(Z  0.4) – P(Z  -0.7)

= 0.6554 – 0.2420
= 0.4134

c) P(Z > 0.73) = 1 – P(Z  0.73)

= 1 – 0.7673
= 0.2327

d) P(Z  k) = 0.9798
k = 2.05 (From tables/CAS calc)

e) P(Z > k) = 0.83

P(Z  k) = 1 – 0.83
= 0.17
 k  -0.95

f) P(-1.2  Z  k) = 0.642
P(Z  k) – P(Z  -1.2) = 0.642
P(Z  k) = 0.642 + P(Z  -1.2)
= 0.642 + 0.1151
= 0.7571
 k  0.70
 - 45 -
g) P(-k  Z  k) = 0.68
P(0  Z  k) = 0.34
P(Z  k) = 0.34 + P(Z  0)
= 0.34 + 0.5
= 0.84
k  0.99

Ref: Ex.10A Q.1-17 (odd)

If the continuous random variable X is normally distributed with a mean m and

standard deviation s and hence variance s2, then it is written as X~N(m,s2), where
m and s2 are the parameters. As it is unlikely that a distribution will have m = 0 and
s = 1, i.e. be a standard normal distribution, then it would be necessary to
standardize the scores. For X, a normally distributed variable, the proportion of the
distribution that falls below a value of k may be referred to as a percentile or quantile,
e.g. P(Z < k) = 0.7, k is the 70th percentile or 0.7 quantile.

E.g.2. The pulse rates of 17 year old boys is recorded at rest and it is found that the
results are normally distributed with a mean of 76 beats per minute (b.p.m.) and a
standard deviation of 5 b.p.m.
a) Find the probability that a boy’s pulse rate is less than 80 b.p.m.
b) What proportion of boys have a pulse rate between 70 b.p.m. and 80 b.p.m.?
c) What pulse rates do the boys with the lowest 10% of pulse rates have?
d) In a group of boys with pulse rates under 80 b.p.m., what percentage will have
pulse rates greater than the mean?

P(X < 80) = P(Z < 0.8)

= 0.7881

P(70 < X < 80) = P(-1.2 < Z < 0.8)

= 0.7881 – 0.1151
= 0.673

c) P(X < c) = 0.1

i.e. P(Z < k) = 0.1
k  -1.28

c = 69.6
The boys with the lowest 10% of pulse rates have 69.6 b.p.m.
 - 46 -
d) P(76 < X <80 | X < 80) = P(0 < Z < 0.8 | Z < 0.8)

 0.3656
 About 37% of boys with a pulse rate under 80 b.p.m. will have pulse rates greater
than the mean.

Ref: Ex.10B Q.1-20 (even)

 - 47 -
FUNCTIONS AND GRAPHS

1. OPTIMUM VALUE: An optimum value is the best value in a particular situation –

best usually means largest or smallest. Optimisation is the process of finding minimum
and/or maximum values. A table of values is often very useful when trying to find the
optimum value to a particular situation.

E.g.1. Rowan, who owns a goat, is planning to grow vegetables but these must be
fenced off. He has 16 metres of suitable fencing wire. What is the maximum
rectangular area he can enclose given that an existing fence can be used as one
side of the rectangle? Would a semicircle enclose a bigger area? and if, so what
is the difference in area?

x (m) 16 – 2x (m) AREA (m2)

1 14 14
2 12 24
3 10 30
4 8 32
5 6 30
3.9 8.2 31.98
4.1 7.8 31.98

 Maximum rectangular area is 32 m².

½C = 16
C = 32 m
C = 2πr

 5.09 m

A = ½πr²
 40.74 m²
 Yes, a semi-circular area would be bigger by approximately 8.74 m2.

An alternative to “zooming in” on the optimum value using a table values is to graph
the points or rule, then use the graph to determine where the optimum solution occurs.

E.g.2. The average cost of constructing an apartment block of x storeys is given by

where C is in millions of dollars. Determine the optimum

number of storeys to build to minimize the cost.
 - 48 -

From the CAS calc., the minimum is at (41.18,14.30).

 The optimum number of storeys to build is 41 for a cost of $14.3 million.

Ref: Ex.1A Q.1-17 (odd)

2. GRADIENT OF A CURVE: Consider this graph –

Point A is the global minimum point.
Point B is a local maximum point.
Point D is a local minimum point.
Point G is the global maximum point.

From A to B the gradient is positive and hence we say that

the function is increasing over this interval.
The function is decreasing from B to D (negative gradient)
and then increases again from D to G. If the graph continued,
then as x ® ¥, y ® ¥ and as x ® -¥, y ® -¥.

Points C and E are points of inflection and

Point F is a point of horizontal inflection.

The terms concave down and concave up are often used to describe the shape of a graph
near maxima or minima. Points on the curve where it changes from being concave down
to concave up, or vice versa, are called points of inflection. If a graph is momentarily
horizontal at a point of inflection, then the point is a point of horizontal inflection.

Local maxima and minima are sometimes referred to as turning points. Turning points
and points of horizontal inflection can also be referred to as stationary points.

As we pass through a maximum point the gradient changes from positive to negative.
Likewise as we pass through a minimum point the gradient changes from negative to
positive. Hence the gradient is zero at both maxima and minima. For points of
horizontal inflection, whilst the gradient is momentarily zero, no change of sign is
involved.

Ref: Ex.1B Q.1-5

 - 49 -
DIFFERENTIATION

1. GRADIENT: Calculus is a study of motion or growth and deals mainly with the rates at
which continuous changes occur. It provides a very useful method for finding how
rapidly a quantity is changing at any moment of time, i.e. the instantaneous rate of
change. Graphically, the rate of change of one variable (y) with respect to another
variable (x) is the gradient of the graph of the relationship.

If the relationship is linear (i.e. uniform) then the rate of change is constant. If the
relationship is not linear (i.e. not uniform) then the gradient is not constant. So the
gradient at a particular point is required. For graphs which are curved, the rate of
change is indicated by the changing slope of the curve.

To determine the gradient at a point on a non-uniform curve, the average rate of change
between two points on the curve, at progressively smaller intervals, is a good
approximation. This is determined by calculating the gradient of the chord over a very
short interval i.e. at x and x + 0.001. The value which the gradient is approaching as the
points get closer and closer together is called the limit of the gradient. Hence, the
gradient between any two points P(x2,y2) and Q(x1,y1), as Q approaches the fixed point P

is .

E.g.1. Determine the slope of y = x2 at (3,9).

Point P Point Q Gradient of chord PQ

(3,9) (4,16)

(3,9) (3.5,12.25)

(3,9) (3.1,9.61)

(3,9) (3.01,9.0601)

(3,9) (3.001,9.006001)

 As y → (3,9), m → 6.

Ref: Ex.2A Q.1-3

LIMITING CHORD
 - 50 -
2. GRADIENT FUNCTION: The procedure used in Example 1 is called the limiting
chord process. In general, to determine the gradient of the tangent at a point P on a
curve y = f(x) – we choose some other point on the curve, Q, whose x co-ordinate is a
little more than that of P, i.e. at x + h.

Thus the gradient of PQ is , i.e. , and hence

the gradient (of the tangent) at P(x,f(x)) is . This gives the

instantaneous rate of change of the function at P, and differentiation is the process of
determining the gradient function.

E.g.2. Find a rule for the (instantaneous) gradient of the function f(x) = 3x2. Hence
calculate the gradient at the point (3,27).

= 6x
 At (3,27), Gradient = 6(3)
= 18

Ref: Ex.2B Q.1-3

3. DIFFERENTIATION: Thus, when we differentiated 3x2 (in Example 2) with respect

to (the variable) x, we obtained the gradient function 6x. Hence 6x is called the
derivative of 3x2. The derivative or the derived function of f(x) with respect to x is


usually denoted by one of these symbols .

 - 51 -
There are certain short-cuts to the process of differentiating polynomial function, axn.
CONSTANT RULE: If f(x) = c, then f '(x) = 0.
e.g. If f(x) = 7, then f '(x) = 0.

POWER RULE: If f(x) = xn, then f '(x) = [Link]-1.

e.g. If f(x) = x5, then f '(x) = 5x4.

CONSTANT MULTIPLE RULE: If f(x) = k.g(x), then f'(x) = k.g'(x).

e.g. If f(x) = 3x5, then f '(x) = 3 ´ 5x4 = 15x4

Thus, in general, to differentiate a polynomial function of the type –

where n is a non-
n n-1 n-2
negative integer and an, an-1, an-2, … are coefficients of x , x , x , …, we “multiply by
the power and decrease the power by one”. The highest power of x in any polynomial
is called the order of the polynomial.

E.g.3. Find the derivative of:

a) y = 4x3
b) y = -3x2
c)
d) g(x) = 2x4

a) y = 4x3

b) y = -3x2

c) = x⅓
f '(x) = ⅓  x-⅔

d) g(x) = 2x4
g'(x) = 4  2x3
= 8x3

Ref: Ex.2C Q.1-52 (R.H.S.); 53-59 (odd)

When differentiating polynomials of the type y = f(x) ± g(x), there is also a short-cut.
SUM RULE: If f(x) = g(x) + h(x), then f'(x) = g'(x) + h'(x).
 - 52 -
7 6
e.g. If f(x) = 2x + 3x, then f '(x) = 14x + 3.
 - 53 -
E.g.4. Find the derivative of:

a)
b) y = (x – 3)(x – 2)

= 7x² – x-1
y' = 14x – (-1)x-2

b) y = (x – 3)(x – 2)
= x² – 5x + 6

Ref: Ex.2D Q.1-25 (odd)

In the previous example, we multiplied out the brackets before differentiating. If we are
required to differentiate a function such as y = (1 – 2x2)5, then to expand the bracket and
differentiate would be lengthy and liable to error. A method which would simplify the

process is required. The Chain Rule states if y = [f(x)]n, then .

For products, it is often ‘easier’ to differentiate first, then perform the product(s), and
finally, simplify the answer, where appropriate. The Product Rule states that if

y = u(x).v(x), then . Similarly, if y = u(x).v(x).w(x), then

Likewise, the Quotient Rule states that if , then

.
 - 54 -
E.g.5. Differentiate each of the following:
a) y = (5x – 3)(4x + 7)
b)
c) g(x) = (3x – 1)(2x3 – x)7
d) y = 2x(x3 – 5)(1 + x3)

e)

f)

g)

a) y = (5x – 3)(4x + 7)

= 20x + 35 + 20x – 12
= 40x + 23

f '(x) = (2x – 3).½x-½ + x½.2

c) g(x) = (3x – 1)(2x3 – x)7

g'(x) = (2x3 – x)7.3 + (3x – 1)7(2x3 – x)6(6x2 – 1)
= 3(2x3 – x)7 + 7(3x – 1)(6x2 – 1)(2x3 – x)6

d) y = 2x(x3 – 5)(1 + x3)

= 2(x3 – 5)(1 + x3) + 6x3(1 + x3) + 6x3(x3 – 5)

= 2(x6 – 4x3 – 5) + 6x3 + 6x6 + 6x6 – 30x3
= 2x6 – 8x3 – 10 + 6x3 + 6x6 + 6x6 – 30x3
= 14x6 – 32x3 – 10
 - 55 -

e)

f)

g)

Ref: Ex.2E Q.1-19 (odd); 20-23

 - 56 -
APPLICATIONS OF DIFFERENTIATION

1. RATES OF CHANGE: Remember – the derivative measures the rate at which any
dependent variable changes as the independent variable changes.

E.g.1. Scientists have found a crystal which “grows” in the shape of a sphere when
placed in a special solution. The volume, V cm³, of the sphere is increasing in
such a way that at time t seconds it is given by V = 7500 + 3600t – 150t2 for
0 £ t £ 12.
a) Calculate the volume when t = 12.
b) Find an expression for the rate of change of the volume with respect to time.
c) Calculate the rate at which the volume is increasing (in cm 3/s) when t = 2 and
when t = 10.
d) Comment on the rate of change of the volume from t = 2 to t = 10.

a) V = 7500 + 3600t – 150t2

V(12) = 7500 + 3600(12) – 150(12)2
= 29 100 cm³

d) The rate at which the volume is changing (increasing) is slowing as time increases
from t = 2 to t = 10.

Ref: Ex.3A Q.1-10 (even), 11-17 (odd)

2. LOCATING STATIONARY POINTS: As we pass through a maximum point the

gradient changes from positive to negative. Likewise, as we pass through a minimum
point the gradient changes from negative to positive. Hence, the gradient is zero at
both maxima and minima. For points of horizontal inflection, whilst the gradient is
momentarily zero, no change of sign is involved.
 - 57 -
Thus, a stationary point can be determined by f '(x) = 0, and hence the point is a
maxima if f "(x) < 0, a minima if f "(x) > 0, or a point of horizontal inflection if
f "(x) = 0, but not all points of inflection are stationary points. Non-stationary points
of inflection are (usually) found between a pair of turning points.

A graphics calculator will give the location, or a good approximation of the exact
location, of almost all of the noteworthy features of a graph – the x- and y-intercepts,
the local/global maxima and minima, the extreme values (i.e. as x ® ± ¥). The only
noteworthy feature not available on the graphics calculator is any points of inflection –
horizontal or not.

This can be overcome by considering the graph of f "(x) and the points for which
f "(x) = 0, or the stationary points of the graph of f '(x). Alternatively this can be
overcome by solving f "(x) = 0.

As the display on a graphics calculator is limited, we cannot always be sure that we

are seeing the “whole picture”. We need to be able to check the information by calculus
methods. This is especially true when the calculator indicates that there are no roots or
fewer roots than anticipated. For example:
one root is expected, or three roots are expected.

E.g.2. For y = x3 – 9x2 + 15x + 2, use differentiation to determine the nature and
location of any stationary points.
 - 58 -

y = x3 – 9x2 + 15x + 2

Stationary Points occur when

3x2 – 18x + 15 = 0
x2 – 6x + 5 = 0
(x – 1)(x – 5) = 0
x = 1 or x = 5
f(1) = 13 – 9(1)2 + 15(1) + 2 = 9  (1,9)
f(5) = 53 – 9(5)2 + 15(5) + 2 = -23  (5,-23)

[To determine the nature of the stationary points we could either use the second
derivative or consider the gradient on either side of the stationary points.]

OR

Consider the gradient on either side of x = 1 and x = 5:

x = 0.9 x = 1 x = 1.1 x = 4.9 x = 5 x = 5.1
+ve 0 -ve -ve 0 +ve

 (1,9) is a maximum and (5,-23) is a minimum.

Ref: Ex.3B Q.1-10 (even); 11, 12

3. APPLICATIONS: Thus, optimum values can be determined by calculus methods, i.e.

maximum/minimum value when , or graphing the function to be optimized, or

refining the solution using a table of values.

E.g.3. An engineer finds that her profit from manufacturing a certain machine is given
by P(x) = 500 – 324x + 36x2 – x3, where x is the number of machines
manufactured per hour. Using calculus methods, find the number of machines
that should be produced for maximum profit.
 - 59 -
Graphical Method:
P(x) = 500 – 324x + 36x2 – x3
From the CAS calculator, the maximum occurs at 18
& from the graphics calculator, the maximum occurs at 18.000000814...

Calculus Method:
P(x) = 500 – 324x + 36x2 – x3

-3x2 + 72x – 324 = 0

x2 – 24x + 108 = 0
(x – 6)(x – 18) = 0
x = 6 or x = 18

 The company should manufacture 18 machines per hour for a maximum profit of $500.

Ref: Ex.3C Q.1-12 (even)

 - 60 -
ANTIDIFFERENTIATION

1. ANTIDIFFERENTIATION: Antidifferentiation, as the name implies, is the reverse

of differentiation, e.g. If f '(x) = 2x, then f(x) = x2 or f(x) = x2 + 1 or f(x) = x2 – 3, etc. All
of these, and more, have f '(x) = 2x. These are known as a Family of Curves. Thus, we
say that the antiderivative (or primitive) of 2x is x2 + c, where c is a constant. Hence
any function has an infinite number of antiderivatives which differ from each other by
a constant only. Additional information may make it possible to determine the value of
c.

Previously, to differentiate axn we “multiplied by the power and decreased the power

by one”, e.g. If y = axn, then . So, to reverse the process we “increase the

power by one and divide by the (new) power”, e.g. If , then

. Thus, in general, if , then for n ¹ -1.

NOTE: The restriction placed on n for the above rule is because, so far, we do not know

any polynomial, y, for which .

E.g.1. Find the antiderivative of:

a) 2x2
b) x2 + 1
c) (x + 2)(x – 1)

= ⅔x3 + c
 - 61 -

= ⅓x3 + x + c

= ⅓x3 + ½x2 – 2x + c

Given x = -1, y = -1
 - 62 -

= (x½ – x-1)2
= (x – 2x-½ + x-2)

Ref: Ex.4A Q.1-37 (odd)

2. FUNCTION FROM RATE OF CHANGE: Previously we looked at real life situations

where the rate of change (of a function) was required, e.g. rate of change of revenue,
rate of change of volume, etc. Thus antidifferentiation, with some additional
information to determine the particular function, can be used to find the function given
the rate of change.

E.g.2. The rate of change of the volume of a certain spherical crystal was found to be

such that = 36 – 3t mm3/s, t seconds after it was placed in a special

“growing” solution. Find the volume function V(t) and determine the volume 12
seconds after immersion if the crystal had an initial volume of 25 mm3.

 V =  (36 – 3t) dt

= 36t – 1.5t2 + c

But when t = 0, V = 25
36(0) – 1.5(0)2 + c = 25
c = 25
 V = 36t – 1.5t + 25
2

So, when t = 12,

V(12) = 36(12) – 1.5(12)2 + 25
= 241 mm3
 12 seconds after immersion the crystal has a volume of 241 mm3.

Ref: Ex.4B Q.1-1 (odd)

 - 63 -
3. INDEFINITE INTEGRAL: Antidifferentiation is also known as integration and is

denoted by the integral sign . Hence, . So, the antiderivative is

also referred to as the Indefinite Integral or the Primitive Integral as it involves the
constant of integration, + c.

E.g.3. Integrate:
a) 4x3
b) 5 + 3x – 2x2
c) (4x – 3)2

a)
= x4 + c

b)

c)  (4x – 3)2 dx =  (16x2 – 24x + 9 ) dx

Ref: Ex.4C Q.1-15 (odd)

 - 64 -
PATTERNS

1. TESTING CONJECTURES: Mathematical statements can be always true, e.g. if

you multiply two odd numbers you get an odd number; always false, e.g. if you add 1 to
an odd number you get an odd number; or sometimes true and sometimes false, e.g. if
you double a number you get an even number. Statements that are always true are
considered to be fundamental truths.

To test the validity of a generalised conjecture – systematically check specific

examples to see if the statement holds true; the more cases tested, the more confident
we are in the validity of the statement; however, only one counter example is required
to disprove a statement, or produce a formal proof.

A framework for a formal proof involves defining all the conditions, making
statements with reasons, if appropriate, logically sequencing an argument, and making
a concluding statement.

Algebraically, for n being an integer – odd numbers are 2n + 1 or 2n – 1, even numbers

are 2n, consecutive even numbers are 2n – 2, 2n, 2n + 2, 2n + 4, etc., consecutive odd
numbers are 2n – 1, 2n + 1, 2n + 3, etc., integers differing by two are n and n + 2 or
n – 1 and n + 1, and two digit numbers are 10a + b where a is the tens digit and b is the
units digit.

E.g.1. Prove the validity of each of the following statements:

a) One less than the square of any counting number greater than 2 is always a
multiple of 5.
b) The product of two even numbers is always even.

a) If the number is 3, then 32 – 1 = 8; this is not a multiple of 5.

 One less than the square of any counting number greater than 2 is not always a
multiple of 5.

b) Let 2k and 2p be the two even numbers.

 2k  2p = 4kp
= 2(2kp)  Any integer multiple of 2 is even.
 The product of two even numbers is always even.

Ref: PROVING GENERALISATIONS Q.1-10 (even)

2. GEOMETRIC REASONING: A geometric deductive proof consists of a list of

statements, and the reasons why these statements are true. Writing a proof consists of
drawing a diagram, if appropriate, of the given situation, any extensions
made/necessary to the diagram, stating any given information, and what is required to
prove, statements with reasons and concludes with the statement that has been proved,
often finishing with Q.E.D. if no reason required.
 - 65 -
E.g.2. Using this diagram, prove DXOL @ DYOL.

GIVEN: XL = YL; sXLO = sYLO = 90

REQUIRED TO PROVE: XOL  YOL

PROOF:
STATEMENTS REASONS
In XOL and YOL,
XL = YL Given
OL = OL Same line segment
sXLO = sYLO = 90 Given
 XOL  YOL S.A.S.

Ref: GEOMETRIC PROOFS Q.1-8 (even)

 - 66 -
BIVARIATE DATA

1. CORRELATION: Bivariate data is data collected concerning two variables, resulting

in pairs of variates. Bivariate data may be represented graphically by a scattergraph,
scatter plot or a scatter diagram. The independent variable is graphed on the
horizontal axis and the dependent variable on the vertical axis.

A scattergraph will indicate both the strength and the direction of the linear
relationship or correlation between the variates. [See scatter plots on p.67.] If a linear
relationship appears to be appropriate then a line of best fit can be drawn on the
scattergraph. The equation of the line ŷ = ax + b and can be determined from a
calculator with a linear regression mode. This is used to make predictions provided the
linear relationship is reasonably strong.

The correlation coefficient (or Pearson’s correlation coefficient) (r or rxy) measures of

how closely the relationship between two variables approximates a linear relationship,
and -1  r  +1, where -1 is a perfect linear relationship with a gradient of -1, and +1 is
a perfect linear relationship with a gradient of +1. As a general guide:

r STRENGTH OF LINEAR RELATIONSHIP

1 to 0.9 Strong
0.8 to 0.6 Medium Positive
0.5 to 0.3 Weak
0.2 to -0.2 No Significant linear relationship
-0.3 to -0.5 Weak
-0.6 to -0.8 Medium Negative
-0.9 to -1 Strong

Predictions can be either interpolation or extrapolation. Interpolation involves

estimating or predicting within the range of the data. Extrapolation goes beyond known
scores, is generally unreliable and can be quite misleading.

Either the covariance or the correlation coefficient may be used when determining if a
relationship between two sets of variables exists. The covariance (Sxy) is the average of
the product of the deviations of each x score from and each y score from and is

defined as or . The correlation coefficient is

the standardized covariance and is defined as where -1  r  1. Hence,

the covariance could also be defined as Sxy = rxy  sx  sy. The covariance (Sxy) is rarely
used since the advent of calculators.
 - 67 -
A line of best fit or the least squares regression line is a predictor of the y values in a
scattergraph. A common notation for a prediction/regression line is ŷ = ax + b. The

values for a and b can be determined by and . The least squares

regression line passes through .

The residuals are the errors of prediction or the deviations of the observations from the
prediction line. Each residual = actual plotted y value – predicted y value from the
line drawn, i.e. .

Time series data is, literally, data collected over time. Time is always on the horizontal
axis.

E.g.1. The following data gives the marks scored on 7 tests in Mathematics and
Accounting by a student over the course of one year:

MATHEMATICS (m) 22 45 35 55 50 67 70
ACCOUNTING (a) 17 40 45 51 50 54 65

a) Draw a scattergraph for this data.

b) Give the equation of the line of best fit.
c) Use the line of best fit to predict the student’s next test result in Accounting if
the student scored 63 in the next Mathematics test.

a)

b) â  0.80m + 6.58

c) From CAS calc., â(63)  57

 the expected Accounting mark is 57 for a Mathematics mark of 63.

Ref: Ex.5A Q.1-18 (even)

 - 68 -
2. OUTLIERS: An outlier is a data point which is anomalous compared to the other data
points. It stands away from the rest of the data. Outliers can significantly alter the
position of the line of best fit. A line of best fit summarizes the overall trend of the
data and, so it is usually better to omit any outliers. However, this should not be
considered the norm for all statistical work. Removing an outlier or group of outliers is
sometimes called trimming or cropping the data, and ‘harvesting’ the rest of the data.

3. RESIDUALS: A scattergraph will often indicate whether a linear relationship exists or

if some other type of relationship would be better suited to the data. Either the graph
of the data is clearly in a shape other than a line or the graph of the residuals shows an
“imbalance” of data points on one side of the regression line. Your CAS calculator
allows you to test, and use, a number of different regression types.

E.g.2. Draw a graph showing the residuals for the data from E.g. 1.

MATHEMATICS (m) 22 45 35 55 50 67 70
ACCOUNTING (a) 17 40 45 51 50 54 65

Ref: Ex.5B Q.1-7 (odd); 6

 - 69 -
MOVING AVERAGES AND SEASONAL EFFECTS

1. MOVING AVERAGES: The line graph of a time series is usually characterized by

fluctuations and irregularities. These fluctuations generally indicate seasonal variations
within a cycle. One method to “smooth out” the graph of the original time series data is
to find and graph the appropriate moving average. The number of cycles or seasons per
year usually determines the “size” of the moving average. If there are no cycles/seasons
then a 3-point moving average is commonly used. A 3-point average is when 3
successive data points are taken, at a time, and the average for each group of 3 is
calculated.

When working with data with a cycle of 4, 6, ..., the moving average is not ‘level’ with
any score. The average is calculated similarly to a 3-point average.

E.g.1. Nursing staff in a hospital are rostered on three shifts: morning (from 7 a.m. to
3:30 p.m.), evening (from 3 p.m. to 11:30 p.m.) and night (from 11 p.m. to 7:30
a.m.).
a) Calculate the 3-point moving average.
b) Graph the pattern of staff numbers and the 3-point moving averages.
The numbers of staff rostered over a 5 day period were:

DAY SHIFT NUMBER

MORNING 7
MONDAY EVENING 5
NIGHT 3
MORNING 11
TUESDAY EVENING 9
NIGHT 4
MORNING 13
WEDNESDAY EVENING 10
NIGHT 5
MORNING 12
THURSDAY EVENING 8
NIGHT 4
MORNING 8
FRIDAY EVENING 4
NIGHT 2
 - 70 -
DAY SHIFT NUMBER 3-P.M.A.
MORNIN 7 ---
G 5 5
MONDAY
EVENING 3 6⅓
NIGHT
MORNIN 11 7⅔
G 9 8
TUESDAY
EVENING 4 8⅔
NIGHT
MORNIN 13 9
G 10 9⅓
WEDNESDAY
EVENING 5 9
NIGHT
MORNIN 12 8⅓
G 8 8
THURSDAY
EVENING 4 6⅔
NIGHT
MORNIN 8 5⅓
G 4 4⅔
FRIDAY
EVENING 2 ---
NIGHT

Ref: Ex.6A Q.1-9

2. MAKING PREDICTIONS: Many sequences of observations (time series) show a

consistent, long term pattern or trend combined with shorter periodic or cyclic
variations. These cyclic variations could be daily or seasonal. In general, a time series
combines the trend with cyclic or seasonal factors and possible random variations.

The seasonal or cyclic component in a time series can be identified by using the
residuals. In regression calculations, the residual R = actual value – predicted value. In
 - 71 -
time series, the residual R = actual value – the moving average or expected value. In
both cases, the residual is the difference between the actual and expected values.
 - 72 -
E.g.2. Calculate the residuals, R, for the nursing staff numbers in E.g.1.

DAY SHIFT NUMBER 3-P.M.A. RESIDUAL

MORNING 7 --- ---
MONDAY EVENING 5 5 0
NIGHT 3 6⅓ -3⅓
MORNING 11 7⅔ 3⅓
TUESDAY EVENING 9 8 1
NIGHT 4 8⅔ -4⅔
MORNING 13 9 4
WEDNESDAY EVENING 10 9⅓ ⅔
NIGHT 5 9 -4
MORNING 12 8⅓ 3⅔
THURSDAY EVENING 8 8 0
NIGHT 4 6⅔ -2⅔
MORNING 8 5⅓ 2⅔
FRIDAY EVENING 4 4⅔ -⅔
NIGHT 2 --- ---

To “seasonally adjust” (or deseasonalize) data, subtract the seasonal residuals from
the original numbers to remove the seasonal or cyclic variations. When a moving
average is used for identifying a cycle, the length of the cycle should be the same as the
length of the moving average. For example, the nursing staff roster the cyclic variations
result from always having most staff on morning shift and least staff on night shift.

E.g.3. Calculate the seasonally adjusted figures for the nursing staff numbers from
E.g.1.

⅓ ⅔ ⅔

⅔ ⅔

⅓ ⅔ ⅔
 - 73 -
DAY SHIFT NUMBE 3-P.M.A. RESIDUA ADJUSTE
R L D
7
MORNIN 7 --- --- 3 /12
G 5 5 0 4.8
MONDAY
EVENING 3 6⅓ -3⅓ 6⅔
NIGHT
MORNIN 11 7⅔ 3⅓ 77/12
G 9 8 1 8.8
TUESDAY
EVENING 4 8⅔ -4⅔ 7⅔
NIGHT
MORNIN 13 9 4 97/12
WEDNESDA G 10 9⅓ ⅔ 9.8
Y EVENING 5 9 -4 8⅔
NIGHT
MORNIN 12 8⅓ 3⅔ 87/12
G 8 8 0 7.8
THURSDAY
EVENING 4 6⅔ -2⅔ 7⅔
NIGHT
MORNIN 8 5⅓ 2⅔ 47/12
G 4 4⅔ -⅔ 3.8
FRIDAY
EVENING 2 --- --- 5⅔
NIGHT

To predict a future value, the trend line is used to find the predicted value and then the
appropriate seasonal residual is added to this.

E.g.4. Give the equation of the regression line for the nursing staff numbers from E.g.1.
Use this equation, together with the seasonal component for the Morning shift
calculated in E.g.3, to predict the number of nursing staff that should be rostered
for Saturday morning.

The trend line equation is ŷ  -0.064x + 7.897

For Saturday Morning shift, x = 16
ŷ(16)  6.872
 Predicted staff numbers  6.87 + 3.4
 10

When working with data with a cycle of 4, 6, ..., it is better to use the centred moving
average. This is calculated by averaging, a pair at a time, the 4-point averages. This is
known as a centred 4-point moving average.
 - 74 -
E.g.5. Shown below are the number of service calls received by an air-conditioning
firm.
a) Calculate the centred 4-point moving averages, the seasonal residuals and the
seasonally adjusted series.
b) Graph, on the same axes, the raw data, the centred 4-point moving averages
and the seasonally adjusted numbers.
c) Use the trend line formed by the moving averages and the seasonal residuals to
predict the service calls for 1992.

1989 1990 1991

FIRST QUARTER 19 23 28
SECOND QUARTER 3 5 7
THIRD QUARTER 5 5 9
FOURTH QUARTER 13 19 21

a)
QUARTER PERIOD CALLS C.4.P.M.A. RESID. SEASONALLY
(x) (y) ADJUSTED
CALLS
1989: 1 1 19 ---- --- 6.875
2 2 3 ---- --- 11.125
3 3 5 10.5 -5.5 12.0625
4 4 13 11.25 1.75 9.875
1990: 1 5 23 11.5 11.5 10.875
2 6 5 12.25 -7.25 13.125
3 7 5 13.625 -8.625 12.0625
4 8 19 14.5 4.5 15.875
1991: 1 9 28 15.25 12.75 15.875
2 10 7 16 -9 15.125
3 11 9 ---- --- 16.0625
4 12 21 ---- --- 17.875

CALL
4.P.M.A.
Seasonal Residuals: st
1 Quarter is = 12.125 S
19 ---
3 10
2nd Quarter is = -8.125 5 11
13 11.5
rd
3 Quarter is = -7.0625 23 11.5
5 13
5 14.25
4th Quarter is = 3.125 19 14.75
28 15.75
7
16.25
9 ---
21
 - 75 -
b) Calls

30

25
Calls

20 C.4-P.M.A.
Seasonally Adjusted
15
Regression line
10

1 2 3 4 5 6 7 8 9 10 11 12 Period

 Seasonally Adjusted Predictions for 1992 are:

1st Quarter: 18.44 + 12.125  31
2nd Quarter: 19.26 - 8.125  11
3rd Quarter: 20.08 - 7.0625  13
4th Quarter: 20.90 + 3.125  24

NOTE: Each centred 4-point moving average is actually based on 5 scores and can be
found directly.

It is the average of half the first score, all of the next three scores and half of the fifth
score, i.e. C.4-P.M.A. = ½x1 + x2 + x3 + x4 + ½x5
4

Ref: Ex.6B Q.1-8 (even)

 - 76 -
LINEAR PROGRAMMING

1. WRITING RESTRICTIONS: Many industrial and economic decisions are based on

the need to optimize some input/output (such as maximize profits by keeping costs down
and revenue up). Linear programming is a technique developed to solve these
problems of planning and management by using linear inequalities.

Solving such problems involves a number of steps –

 Define variables to represent each item.
 Write restrictions in terms of the variables.
 Graph the restrictions and determine the feasible region.
 Define the objective function to be optimized.
 Determine which point(s) in the feasible region give the optimum situation.

Defining the variables and writing the restrictions are the two most important steps
towards solving the problem. Writing the restrictions involves translating the (English)
words/sentences into algebraic symbols/inequalities.

E.g.1. A lolly maker is preparing for the Christmas market. Holly has 1800 kg of jubes,
1500 kg of licorice allsorts and 750 kg of fruit drops. She makes these into two
mixtures: Holiday (h), which sells for $1.50 per kilogram and Santa (s), which
sells for $4.00 per kilogram.
Holiday mixture contains 60% jubes, 30% licorice allsorts and 10% fruit drops.
Santa mixture contains 20% jubes, 50% licorice allsorts and 30% fruit drops.
Write each of the following statements in terms of h and s:
a)the weight of jubes possible,
b) the weight of licorice allsorts possible,
c)the weight of fruit drops possible, and
d) the total cost of the Holiday and Santa mixtures.

a) Jubes: 0.6h + 0.2s  1800

b) Licorice allsorts: 0.3h + 0.5s  1500

c) Fruit drops: 0.1h + 0.3s  750

d) Cost = $(1.5h + 4s)

Ref: Ex.7A Q.1-6, 12-18, 24-27, 32-36, 41-45

2. GRAPHING INEQUALITIES: To graph an inequality, graph the appropriate

equation, test a point not on the line, usually (0,0), and shade the "correct" section. When
graphing inequalities, remember that a graph of < or > has a broken line to indicate that
this is the boundary but is not included in the region whereas  or  are bounded by a
solid line.
 - 77 -
E.g.2. Graph each of the following inequalities
a) x  2,
b) y  2x – 1, and
c) y > -x – 2.

a) Graph: x = 2  a solid.
Shading: x  2 is to the left of x = 2.

b) Graph: y = 2x – 1  a solid line.

Shading: y  2x – 1 does not contain (0,0).

c) Graph: y = -x – 2  a broken line.

Shading: y > -x – 2 does contain (0,0).
 - 78 -
E.g.3. The graph the restrictions d  0, s  0, s  d + 3, s  3 – d and d  4.

Ref: Ex.7B Q.1-23 (odd)

3. DETERMINING THE OPTIMUM SOLUTION: When two or more inequalities are

graphed on the same set of axes, the resulting shaded area is called the feasible region or
solution space because it satisfies all the inequalities (constraints). One of the vertices
of the feasible region is (usually) the optimum solution. For real life situations, the
constraints x  0 and y  0 are assumed to apply. The input/output to be optimized is
sometimes referred to as the objective function.

E.g.4. For the graph in E.g.3, locate the points which:

a) maximize s,
b) minimize d,
c) maximize 2d + s, or
d) minimize 2d – s.

The points of the feasible region are (0,3), (3,0), (4,0) and (4,7).

(0,3) (3,0) (4,0) (4,7) SOLUTION

Max s 3 0 0 7 (4,7)
Min d 0 3 4 4 (0,3)
Max 2d + s 3 6 8 15 (4,7)
Min 2d – s -3 6 8 1 (0,3)

NOTE: In real life situations a solution of -3 may not be practical and so the next best
solution would be used, i.e. 2d – s is minimized at (4,7). This solution would
satisfy a question such as maximum income for minimum costs.

Ref: Ex.7C Q.1-10 (even)

 - 79 -
E.g.5. Vivian wishes to make two models of a can-opener, a standard model and a
deluxe model. He can make no more than 50 can-openers per hour. He can make
no more than 20 deluxe models and no more than 40 standard models per hour.
He makes 50 cents profit on a deluxe model and 40 cents profit on a standard
model. What would you recommend the company to produce each hour in order
to gain a maximum profit?

Let d be the number of deluxe model

and s be the number of standard model can-openers produced.

Constraints:
d + s  50 (Total production  50)
d  20
s  40
d  0 and s  0

Graph:

Objective Function: Profit = $0.50d + $0.40s

Feasible Region:

VERTEX PROFIT
(0,0) $0
(0,40) $16
(10,40) $21
(20,30) $22
(20,0) $10

 Maximum profit of $22 occurs when 20 deluxe and 30 standard can-openers are made
per hour.

Ref: Ex.7D Q.1-14 (even)

 - 80 -
AREA UNDER A CURVE

1. AREA UNDER A CURVE: Areas of regular figures such as squares, circles, etc. can be
readily found by common formulae. However, the area of non-regular figures such as the
curve y = f(x) are not so readily found. Some “boundaries” for the area being considered
need to be set. The area under the curve is usually bounded by the curve, the X-axis
and the lines x = a and x = b for an interval [a;b].

One method of approximating the area is to draw an accurate graph on grid paper,
987654321
determine the area of one “square”, then simply 21 count the “squares”, and multiply by
the area of one square.

E.g.1. Use the graph of y = x3 to estimate the area under the curve from x = 1 to x = 2.
y 3
y = x
Area 1 square = 0.1  1 = 0.1 units 2 9
There are approximately 37 little blocks in the 8
7
required region.
6
5
 Area  = 3.7 units 2 4
3
2
1

1 2 x

Another method is to divide the interval [a;b] into a number of equal parts and construct
rectangles, both upper and lower, and then sum the areas of these rectangles (SU and SL).
 - 81 -
The sum of the (upper) rectangular areas (SU) is an overestimate of the required area and
the sum of the (lower) rectangular areas (SL) is an underestimate of the required area, so

the “true” value lies somewhere between the two. Thus, A  . The more sub-

intervals taken, the smaller the difference between SU and SL and

the better the approximation to the true area.

E.g.2. Given the curve y = x2 + 3, divide the interval [1;4] into three sub-intervals and
construct both the upper and lower rectangles on them. Hence approximate the
area bounded by the X-axis and the ordinates x = 1 and x = 4.

Overestimate: Underestimate:
R1 = 1  7 = 7 R1 = 1  4 = 4
R2 = 1  12 = 12 R2 = 1  7 = 7
R3 = 1  19 = 19 R3 = 1  12 = 12
SU = 38 SL = 23

Ref: Ex.8A Q.1-3

 - 82 -
PROJECT NETWORKS

1. CRITICAL PATH: Networks can be concerned with timing and tasks which must be
completed in order. The tasks are shown on a digraph (directed graph) which gives a
possible allocation of tasks. The arcs show the order in which the tasks are completed.
Each node, called an event, represents a stage of completion of all the tasks on the paths
which finish at that node. Since completing the “project” requires that all the tasks are
done, the shortest time for completion is the time taken on the longest path. The path
with the longest time is called the critical path, and any “spare” time is called slack or
float time.

E.g.1. The digraph below shows the times, in minutes, needed to complete various tasks
which make up a project.
a) What is the minimum time required for the completion of all tasks?
b) What is the critical path?
c) Identify slack areas.
20
23 B 1
4
60
34 60
37 F 1 J 8
5
52
20

K
23 52
A

19
C

Start
12

79
G

0
79
Finish
40 22
D

L
18

40
2 H 1
E 2 0
18 50
18 57

a) The minimum time for the whole project is 79 minutes.

b) The critical path is DEGJK.
c) Delays in starting A, B, C, F, H or L could occur without affecting the finishing time.

Ref: Ex.9A Q.1-10 (even)

2. PROJECT NETWORKS: A project network represents a timing/task situation in a

network where any task with no prerequisites gets a line from the start and its time is
written along the line. Tasks with prerequisites are then added, each task drawn as a line
from its prerequisites.
 - 83 -
E.g.2. Three friends are working on a combined assignment made up of 10 activities
(see the table below).

ACTIVITIES A B C D E F G H I J
TIME (hrs) 3 2 3 6 4 8 6 7 3 4
PREREQUISITES - - - A C B B D, F E, G H, I

a) Draw a project network for the given activities.

b) What is the shortest time needed to complete the assignment?
c) Which activities are critical, in that extra time taken on them would result in
delaying the completion of the assignment?
d) Mark the critical path on your diagram.

a) & d)

b) The shortest time is 21 hours.

c) The critical activities are B, F, H and J.

Ref: Ex.9B Q.1-12 (even)