Professional Documents
Culture Documents
Further Math Study Note
Further Math Study Note
HEADSTART LECTURE
JANUARY 2020
Presented by Josh Hamilton
LECTURE PLAN BLOCK 1:
25 minutes
OVERVIEW & STUDY TIPS
Get into good habits early and
know what’s in store for the year
BREAK 1 5 minutes + 5 minutes for
questions
BLOCK 2:
45 minutes
DATA ANALYSIS 1
Content! Fairly boring compulsory content!
+ 10 minutes
BREAK 2 5 minutes
for questions
BLOCK 3:
35 minutes
DATA ANALYSIS 2
Less boring but still compulsory content!
QUESTIONS 10 minutes
2
LECTURE PLAN
Note: • Lecture slides will be are available in the resource
tab below, so don’t stress if you don’t get everything
down!
3
BLOCK 1:
OVERVIEW & STUDY TIPS
4
COURSE OVERVIEW
Matrices SACS
Networks 33% of
MODULES X 2
Geometry Final
Graphs & Linear Relations
Mark
5
WHAT DO I NEED TO KNOW?
Read the
study design!
6
YEAR 12
• First off congrats on surviving 12 years of school - only 1
more to go
7
MOTIVATION IN YEAR 12
By the time
exams come
around most
Most students people start
during term 3 studying way
more
8
DEALING WITH YOUR OWN EXPECTATIONS
• Know what your end game is, being successful in VCE does
not mean you need to get a 90+ ATAR. VCE is just a
steppingstone to get where you want to go, and the most
important thing is that you get there.
• For Further, if you are aiming for the higher end of grades
(e.g. 40+) I would say aim for a 45. This is because there is
very little to distinguish between a 45 and a 50 – it comes
down to whether the questions play to your strengths on the
day.
9
DEALING WITH YOUR OWN EXPECTATIONS
– Change your study habits because what you have been
doing may not have worked thus far.
– Be prepared to make sacrifices throughout the year to
ensure you get the results you want (but maintain a
healthy balance) – sometimes it might take saying that
you can’t go out on a Friday or Saturday night so you can
get your work done. #covid
– Be strict on yourself when you think you are drifting and
lowering your expectations as you think they may not be
attainable and try to stay on course.
– Set small goals to make your larger goal seem less
daunting – my first goal for year 12 was to get an A+ for
my first SAC for every subject.
10
RESOURCES FOR FURTHER
• Calculator
• Summary Book
• Teachers, tutors
11
TEXTBOOK/STUDY GUIDES
12
CALCULATOR
• Calculator guides in textbook – copy them out into your
notes in case you go blank in a SAC or exam!
• Your CAS is your best friend, make sure you know it well
– For my trigonometry friends, you WILL love your CAS cause bearing
are a thing and can be a massive pain – so please learn how to use
it well!
13
SUMMARY BOOK
• There are two (maybe three) models for a summary book
that students follow:
– Student A who puts literally everything he/she sees in their
textbooks/study guides and class notes. At the end of the year their
reference book is THICC but they have the peace of mind that
everything's in there.
• The negatives:
– It will be a pain to look through during SACs or Exams and might waste
valuable time.
– Might consume a lot of your study time making it – if using this method you
NEED to stay up to date!
• The positives:
– You have the peace of mind that you have all the content and can use it
when you get stuck.
– Creating a comprehensive summary book is a good tool to revise content
14
SUMMARY BOOK
15
SUMMARY BOOK
MY ADVICE -
DISCLAIMER: take what I say with a grain of salt as this is from
my experiences and you might be different.
One of the Golden Rules is that you should create your own
reference book (don’t buy one) – you learn so much by
understanding the formulas and not just memorizing how to
answer specific questions.
16
SUMMARY BOOK
The essentials to consider when thinking about your own
summary book….
• Approach with a given mindset – minimalistic,
comprehensive or concise.
17
TEACHERS/TUTORS
• (most of) your teachers aren’t out to get you!
– Utilise their help – they have seen 100’s and 1000’s of students go
through VCE they know what differentiates a 35 from a 40.
– The more you show you are keen and hardworking the more likely
your teachers will out time aside to help you at the end when it gets
super congested for them!
18
OTHER TIPS
• Don’t touch last year’s exam until the end of the year
CODE: JOSHVCE
UNIVARIATE DATA BIVARITE DATA
DISTRUBUTIUONS DISTRUBUTIUONS
DATA ANALYSIS
The biggest part of the course
à Worth the most marks
à Most important?
MODELLING
TIME SERIES DATA
LINEAR
ASSOCIATIONS
23
UNIVARIATE DATA DISTRIBUTIONS
TYPES OF DATA
Univariate Bivariate
1 2
Categorical Numerical
24
UNIVARIATE DATA DISTRIBUTIONS
TYPES OF DATA – TYPES OF VARIABLES
What type of variables?
Categorical Numerical
25
UNIVARIATE DATA DISTRIBUTIONS
TYPES OF DATA – TYPES OF VARIABLES
Categorical Data
Data is divided into categories
e.g. Eye colour, football team
No sensible way to sort the categories! Categories have a natural order, but the
interval between them is not specific.
e.g. Hair colour: e.g. How satisfied are you?
Does it make sense to order blonde, brown, Very Satisfied
red and black hair? No! Somewhat Satisfied
Neutral
Somewhat Dissatisfied
Very Dissatisfied
26
UNIVARIATE DATA DISTRIBUTIONS
TYPES OF DATA – TYPES OF VARIABLES
Numerical Data
Data that you measure or count
e.g. Height, number of students
Data you can count, can only take on a Data that you measure, can take any
finite set of values. value (infinite possibilities)
e.g. Number of people at this lecture! e.g. How much does a $2 coin weight?
I could count all of you, and I would get a 7 grams
distinct number. Even if I wanted, I couldn’t 6.6 grams
get a more ‘accurate’ value. 6.60 grams
6.601 grams
…
27
UNIVARIATE DATA DISTRIBUTIONS
TYPES OF DATA – TYPES OF VARIABLES
What type of variables?
Categorical Numerical
28
UNIVARIATE DATA DISTRIBUTIONS
TYPES OF DATA – CHECKLIST
Can it be manipulated?
YES i.e. does it make sense to NO
find the mean, median,
mode, range, multiply it,
add it?
Numerical Categorical
Numerical Categorical
30
UNIVARIATE DATA DISTRIBUTIONS
TYPES OF DATA – CHECKLIST
NO YES
Numerical Categorical
• For example, numbers on uniforms are used to identify players, they act
as names, and therefore they are categorical.
• Post codes, house numbers and ratings on number scales (e.g. rate out
of 5 stars) are other common categorical variables that use numbers!
31
UNIVARIATE DATA DISTRIBUTIONS
PRACTICE QUESTIONS
Assign the following variables as numerical, categorical, and whether they
are discrete/continuous or ordinal/nominal:
32
UNIVARIATE DATA DISTRIBUTIONS
PRACTICE QUESTIONS
Assign the following variables as numerical, categorical, and whether they
are discrete/continuous or ordinal/nominal:
33
UNIVARIATE DATA DISTRIBUTIONS
PRACTICE QUESTION
Univariate Bivariate
1 2
DATA
What type of variables?
Categorical Numerical
36
UNIVARIATE DATA DISTRIBUTIONS
TYPES OF DATA – NUMBER OF VARIABLES
Univariate
Colour Number
Red 3
Black 10
White 8
Silver 5
37
TYPES OF DATA – NUMBER OF VARIABLES
Frequency tables
Categorical Percentage frequency tables
Bar charts
Frequency Tables
Dot plots
Numerical Box plots
Stem and leaf plots
Histograms
39
UNIVARIATE DATA DISTRIBUTIONS
DISPLAYING UNIVARIATE DATA
Frequency tables
Categorical Percentage frequency tables
Bar charts
Frequency tables
Dot plots
Numerical Box plots
Stem and leaf plots
Histograms
40
UNIVARIATE DATA DISTRIBUTIONS
FREQUENCY TABLES Categorical
41
UNIVARIATE DATA DISTRIBUTIONS
PERCENTAGE FREQUENCY TABLES Categorical
42
UNIVARIATE DATA DISTRIBUTIONS
BAR CHARTS Categorical
100
Frequency
Pets at home
50
0
Cats Dogs Birds Fish
Type of pet
43
UNIVARIATE DATA DISTRIBUTIONS
DESCRIBING CATEGORICAL DISTRIBUTIONS
44
UNIVARIATE DATA DISTRIBUTIONS
EXAMPLE QUESTION
Q: Comment on the data shown in the frequency table below.
Climate Frequency % Frequency
Hot 6 26.1%
Mild 14 60.9%
Cold 3 13.0%
Total 23 100
45
UNIVARIATE DATA DISTRIBUTIONS
DESCRIBING CATEGORICAL DISTRIBUTIONS
Climate Frequency % Frequency
Hot 6 26.1%
Mild 14 60.9%
Cold 3 13.0%
Total 23 100
Context
46
UNIVARIATE DATA DISTRIBUTIONS
DESCRIBING CATEGORICAL DISTRIBUTIONS
Climate Frequency % Frequency
Hot 6 26.1%
Mild 14 60.9%
Cold 3 13.0%
Total 23 100
Mode
47
UNIVARIATE DATA DISTRIBUTIONS
DESCRIBING CATEGORICAL DISTRIBUTIONS
Climate Frequency % Frequency
Hot 6 26.1%
Mild 14 60.9%
Cold 3 13.0%
Total 23 100
48
UNIVARIATE DATA DISTRIBUTIONS
FREQUENCY TABLES Numerical
Age Frequency
17 10
18 13
19 2
Age Frequency
0-20 18
20-40 43
40-60 25
49
UNIVARIATE DATA DISTRIBUTIONS
DOT PLOTS Numerical
1 2 3 4 5
50
UNIVARIATE DATA DISTRIBUTIONS
DOT PLOTS Numerical
Common question:
1 2 3 4 5
51
UNIVARIATE DATA DISTRIBUTIONS
DOT PLOTS Numerical
Common question:
1 2 3 4 5
52
UNIVARIATE DATA DISTRIBUTIONS
STEM AND LEAF PLOTS Numerical
4 1
Stem Leaf
(essentially representing 40) (essentially representing 1)
53
UNIVARIATE DATA DISTRIBUTIONS
STEM AND LEAF PLOTS Numerical
Stem Leaf
1 0 2 10
Key: 2 2 3 8 9
3 9 9
1 | 0 = 10 4 4 4 7
5
6 0 1 9
54
UNIVARIATE DATA DISTRIBUTIONS
STEM AND LEAF PLOTS Numerical
Stem Leaf
1 0 2 100
Key: 2 2 3 8 9
3 9 9
1 | 0 = 100 4 4 4 7
5
6 0 1 9
55
UNIVARIATE DATA DISTRIBUTIONS
STEM AND LEAF PLOTS Numerical
56
UNIVARIATE DATA DISTRIBUTIONS
HISTOGRAMS Numerical
57
UNIVARIATE DATA DISTRIBUTIONS
HISTOGRAMS - LOG SCALES
• Sometimes, the data’s range is too large to display on a
regular histogram
58
UNIVARIATE DATA DISTRIBUTIONS
HISTOGRAMS - LOG SCALES
Properties of logs:
Eg. 5
4.5
3.5
2.5
Frequency
2
1.5
0.5
0
1 2 3 4 5
Log(variable)
60
UNIVARIATE DATA DISTRIBUTIONS
HISTOGRAMS - LOG SCALES
61
UNIVARIATE DATA DISTRIBUTIONS
HISTOGRAMS - LOG SCALES
Use calculator!!
- To find the number of a log
- 101.683 = 48.1948
62
ANOTHER PRACTICE QUESTION
Includes:
65
UNIVARIATE DATA DISTRIBUTIONS
THE 5 FIGURE SUMMARY
2 5 7 8 9 13 14 15 16 20 21 25 37 41
66
UNIVARIATE DATA DISTRIBUTIONS
THE 5 FIGURE SUMMARY
2 5 7 8 9 13 14 15 16 20 21 25 37 41
67
UNIVARIATE DATA DISTRIBUTIONS
THE 5 FIGURE SUMMARY
2 5 7 8 9 13 14 15 16 20 21 25 37 41
Median: 14.5
68
UNIVARIATE DATA DISTRIBUTIONS
THE 5 FIGURE SUMMARY
How can we find the 5 number summary by hand?
2 5 7 8 9 13 14 15 16 20 21 25 37 41
Q1: 8
Median: 14.5
69
UNIVARIATE DATA DISTRIBUTIONS
THE 5 FIGURE SUMMARY
How can we find the 5 number summary by hand?
2 5 7 8 9 13 14 15 16 20 21 25 37 41
Q1: 8
Median: 14.5
Q3: 21
70
70
UNIVARIATE DATA DISTRIBUTIONS
THE 5 FIGURE SUMMARY
How can we find the 5 number summary by hand?
2 5 7 8 9 13 14 15 16 20 21 25 37 41
Minimum: 2
Q1: 8
Median: 14.5
Q3: 21
Max: 41
71
UNIVARIATE DATA DISTRIBUTIONS
THE 5 FIGURE SUMMARY
How can we find the 5 number summary from a dot plot?
1 2 3 4 5
72
UNIVARIATE DATA DISTRIBUTIONS
THE 5 FIGURE SUMMARY
How can we find the 5 number summary from a stem plot?
Stem Leaf
1 0 2
2 2 3 8 9
3 9 9
4 4 4 7
5
6 0 1 9
73
UNIVARIATE DATA DISTRIBUTIONS
THE 5 FIGURE SUMMARY
What can we do with these details?
- IQR
- IQR = Q3 – Q1
74
UNIVARIATE DATA DISTRIBUTIONS
BOX PLOTS Numerical
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
75
UNIVARIATE DATA DISTRIBUTIONS
BOX PLOTS Numerical
Q1 Median Q3
Min Max
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
76
UNIVARIATE DATA DISTRIBUTIONS
BOX PLOTS Numerical
Q1 Median Q3
Min Max
Outlier Outlier
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
77
UNIVARIATE DATA DISTRIBUTIONS
BOX PLOTS Numerical
Q1 Median Q3
Min Max
Outlier Outlier
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
78
UNIVARIATE DATA DISTRIBUTIONS
ANALYSING / DESCRIBING DATA
Outliers: Are there any present? If so, what are they? Also note
if there are no outliers.
79
UNIVARIATE DATA DISTRIBUTIONS
DESCRIBING HISTOGRAMS
Shape:
80
UNIVARIATE DATA DISTRIBUTIONS
DESCRIBING BOX PLOTS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
Approximately Symmetrical
81
UNIVARIATE DATA DISTRIBUTIONS
NORMAL DISTRIBUTION
Symmetrical
50% 50%
Approaches zero
on both sides
82
UNIVARIATE DATA DISTRIBUTIONS
STANDARD DEVIATION
e.g. mean = 5 sx = 2
83
UNIVARIATE DATA DISTRIBUTIONS
68-95-99.7% RULE
• Around 68% of the data values lie within one standard
deviation of the mean.
• Around 95% of the data values lie within two standard
deviations of the mean.
• Around 99.7% of the data values lie within three
standard deviations of the mean.
84
UNIVARIATE DATA DISTRIBUTIONS
PRACTICE QUESTIONS
85
UNIVARIATE DATA DISTRIBUTIONS
PRACTICE QUESTIONS
32– 28 = 4
4/2 = 2
2 std deviations under the mean
34% + 13.5% = 47.5%
24 x 0.475 = 11.4
11 students
86
UNIVARIATE DATA DISTRIBUTIONS
STANDARD DEVIATION
𝑥 − 𝑥̅
𝑧=
𝑠
𝑧 = z-score
𝑥 = actual score
𝑥̅ = mean
𝑠 = standard deviation
87
UNIVARIATE DATA DISTRIBUTIONS
STANDARD DEVIATION
𝑥 = 𝑥̅ + (𝑧 ×𝑠)
Actual score = Mean + (z-score × standard deviation)
88
UNIVARIATE DATA DISTRIBUTIONS
PRACTICE QUESTIONS
Q1. 81.5%
Q2.
89
UNIVARIATE DATA SUMMARY
Univariate Categorical
DATA Types of data
Bivariate Numerical
Frequency tables
Percentage frequency
Categorical tables
Bar charts
Displaying data
Frequency tables
Dot/Box plots
Numerical
Stem and leaf plots
Histograms
DATA ANALYSIS
The biggest part of the course
à Worth the most marks
à Most important?
MODELLING
TIME SERIES DATA
LINEAR
ASSOCIATIONS
91
TYPES OF DATA – NUMBER OF VARIABLES
92
WHY BIVARIATE DATA?
93
BIVARIATE DATA DISTRIBUTIONS
EXPLANATORY VS RESPONSE VARIABLES
94
BIVARIATE DATA DISTRIBUTIONS
EXPLANATORY VS RESPONSE VARIABLES
Explanatory Response
variable variable
Type of Data
Explanatory Graph
Response Variable
Variable
96
BIVARIATE DATA DISTRIBUTIONS
DISPLAYING BIVARITE DATA
Type of Data
Explanatory Graph
Response Variable
Variable
97
BIVARIATE DATA DISTRIBUTIONS
DISPLAYING BIVARITE DATA
Type of Data
Explanatory Graph
Response Variable
Variable
98
BIVARIATE DATA DISTRIBUTIONS
DISPLAYING BIVARITE DATA
Type of Data
Explanatory Graph
Response Variable
Variable
Segmented Bar Chart or
Categorical Categorical Two Way Frequency Table
99
BIVARIATE DATA DISTRIBUTIONS
SEGMENTED BAR CHARTS Categorical Categorical
Year
100
BIVARIATE DATA DISTRIBUTIONS
SEGMENTED BAR CHARTS Categorical Categorical
Year
101
BIVARIATE DATA DISTRIBUTIONS
TWO WAY FREQUENCY TABLE Categorical Categorical
- RV = Rows
- EV = Columns
102
BIVARIATE DATA DISTRIBUTIONS
TWO WAY FREQUENCY TABLE Categorical Categorical
103
BIVARIATE DATA DISTRIBUTIONS
BACK TO BACK STEM & LEAF Numerical Categorical
8 5 Key
1 6 0 2 4 6 9
8 3 0 7 135 1 | 0 = 10
5 4 2 0 8 5
9 6 2 9 9
10 0 0 0
104
BIVARIATE DATA DISTRIBUTIONS
PARALLEL DOT PLOTS Numerical Categorical
Boys
9 10 11 12 13 14 15 16 17
Girls
9 10 11 12 13 14 15 16 17
105
BIVARIATE DATA DISTRIBUTIONS
COMPARING DISTRIBUTIONS
- Box Plots
- Histograms
- Dot Plots
- Back to Back Stem and Leaf Plots
We look at:
- Centre
- Spread
- Shape
106
BIVARIATE DATA DISTRIBUTIONS
COMPARING DISTRIBUTIONS
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26
Score
107
BIVARIATE DATA DISTRIBUTIONS
SCATTERPLOTS Numerical Numerical
20
15
Y
10
0
0 5 10 15 20 25
108
BIVARIATE DATA DISTRIBUTIONS
SCATTERPLOTS Numerical Numerical
25
20
15
Y
10
0
0 5 10 15 20 25
1. Strength
2. Direction
3. Form
109
BIVARIATE DATA DISTRIBUTIONS
PEARSON’S CORRELATION COEFFICIENT
110
BIVARIATE DATA DISTRIBUTIONS
PEARSON’S
PearsonCORRELATION COEFFICIENT
Correlation Coefficient
111
BIVARIATE DATA DISTRIBUTIONS
DIRECTION
• Positive or negative
Positive Negative
112
BIVARIATE DATA DISTRIBUTIONS
FORM
113
BIVARIATE DATA DISTRIBUTIONS
INTERPRETING r
Linear, negative and strong Linear, negative and Linear, negative and weak
It can be concluded that the y moderate There is limited evidence to
variable should decrease as the There is some evidence to suggest that the y variable
x variable increases. suggest that the y variable should decrease as the x
should decrease as the x variable increases.
variable increases.
114
BIVARIATE DATA DISTRIBUTIONS
NON – CAUSAL EXPLANATIONS
115
BIVARIATE DATA DISTRIBUTIONS
CORRELATION OR CAUSATION?
116
BIVARIATE DATA DISTRIBUTIONS
CORRELATION OR CAUSATION?
117
BIVARIATE DATA DISTRIBUTIONS
CORRELATION OR CAUSATION?
118
BIVARIATE DATA DISTRIBUTIONS
CORRELATION OR CAUSATION?
119
BIVARIATE DATA DISTRIBUTIONS
CORRELATION AND CAUSATION
Causation
x y
120
BIVARIATE DATA DISTRIBUTIONS
NON-CAUSAL EXPLANATIONS
• Common response: both variables are linked to a
third, shared variable
x y
Lurking variable
x y
? ?
Confounding variable
121
BIVARIATE DATA DISTRIBUTIONS
NON-CAUSAL EXPLANATIONS
• Coincidence: simply by chance
????
x y
122
BIVARIATE DATA DISTRIBUTIONS
POPULATION AND SAMPLING
Key Terms
123
BIVARIATE DATA DISTRIBUTIONS
POPULATION AND SAMPLING
124
BIVARIATE DATA DISTRIBUTIONS
POPULATION AND SAMPLING
125
BIVARIATE DATA SUMMARY
Type of Data
Explanatory Response
Graph
Variable Variable
Segmented Bar Chart
Categorical Categorical or Two Way Frequency
Table
Explanatory Response
µ, σ vs. 𝑥,̅ 𝑠
variable variable
x y
Population vs Sample Statistics
Variables and causality
126
CODE: JOSHVCE
QUESTIONS
BREAK 2 – 5 Minutes
QUESTIONS
BREAK 2 – 5 Minutes
QUESTIONS
QUESTIONS
BREAK 2 – 5 Minutes
CODE: JOSHVCE
UNIVARIATE DATA BIVARITE DATA
DISTRUBUTIUONS DISTRUBUTIUONS
DATA ANALYSIS
The biggest part of the course
à Worth the most marks
à Most important?
MODELLING
TIME SERIES DATA
LINEAR
ASSOCIATIONS
131
MODELLING DATA DISTRIBUTIONS
MODELLING DATA
• Bivariate data, and in particular bivariate data with two
numerical variables, is extremely useful!
132
MODELLING DATA DISTRIBUTIONS
LEAST SQUARES LINE OF BEST FIT
Few things:
- Least= vertical
We take the residual Squares distanceRegression
between actual data point and line of
best fit .
- We then make sure our line of best fit line minimises the sum of the squares of
• Residual
residuals = vertical distance
between actual data point
- Works best if there are no outliers
and line of best fit
𝑦 = 𝑎 + 𝑏𝑥
• Two scenarios:
1. If you’re given the raw data, use your CAS!
And:
• 𝑟 is the Pearson correlation coefficient
• 𝑠" and 𝑠# are the sample standard deviations of 𝑦 and 𝑥 respectively.
• 𝑥̅ and 𝑦, are the sample means of 𝑥 and 𝑦
135
MODELLING DATA DISTRIBUTIONS
EXAMPLE
Example
136
MODELLING DATA DISTRIBUTIONS
INTERPRETING REGRESSION LINES
137
MODELLING DATA DISTRIBUTIONS
COEFFICIENT OF DETERMINATION
- r2
139
MODELLING DATA DISTRIBUTIONS
RESIDUALS
• How can we mathematically check if a scatterplot is linear
or not?
Linear
141
MODELLING DATA DISTRIBUTIONS
DATA TRANSFORMATION
Circle of transformations
142
MODELLING DATA DISTRIBUTIONS
Transforming data
EXAMPLE
•
Transforming
If we applied anApplying
2
x2 transformation,
data
a x transformation
we would fit a least
squares
y regression
1 line to: 10
4.5 16 26
xy 1 2
4.5 3 10 4
16 5
26
xX^2
2 1 4 9
Transforming 16
data 25
• The
If weregression
applied
Therefore, anline would
x2graph
we would have theequation:
transformation,
and find we would
recurrence fit aofleast
equation
squares
! = # +regression
%& v line to:
y 1 4.5 10 16 26
X^2 1 4 9 16 25
𝑦 𝑦
𝑥 𝑥!
144
MODELLING DATA SUMMARY Circle of transformations
𝑦 = 𝑎 + 𝑏𝑥
g residual plots
Essential Further Mathematics Units 3 & 4, 4th Edition, pg 190.
ws a linear trend
Interpreting and calculating lines of best fit Transformations
CASE 2: Good choice
!
Residual
plots
𝑟
Correlation coefficient
DATA ANALYSIS
The biggest part of the course
à Worth the most marks
à Most important?
MODELLING
TIME SERIES DATA
LINEAR
ASSOCIATIONS
146
TIME SERIES
20
Sea Level Rise (cm)
15
10
0
1883
1887
1891
1895
1899
1903
1907
1911
1915
1919
1923
1927
1931
1935
1939
1943
1947
1951
1955
1959
1963
1967
1971
1975
1979
1983
1987
1991
1995
1999
2003
2007
2011
147
TIME SERIES
20
Sea Level Rise (cm)
15
10
0
1883
1887
1891
1895
1899
1903
1907
1911
1915
1919
1923
1927
1931
1935
1939
1943
1947
1951
1955
1959
1963
1967
1971
1975
1979
1983
1987
1991
1995
1999
2003
2007
2011
148
TREND
Describes what is happening in the long term
GDP of Australia
http://www.abs.gov.au/websitedbs/D3310114.nsf/home/Time+Series+Analysis:+The+Basics
149
SEASONALITY
- Peaks/troughs at regular intervals related to the calendar
Seasonality
(usually seasons of the year, but could be weekly, monthly
http://www.abs.gov.au/websitedbs/D3310114.nsf/home/Time+Series+Analysis:+The+Basics
http://www.abs.gov.au/websitedbs/D3310114.nsf/home/Time+Series+Analysis:+The+Basics 150
CYCLES
http://robjhyndman.com/hyndsight/cyclicts/
151
IRREGULAR FLUCTUATIONS
152
tells us that, even though demand for accommodation has fluctuated from month to month,
demand for hotel and motel accommodation has increased over time.
STRUCTURAL CHANGE
Structural change
Structural
- Where change
there is a sudden change in the established pattern
350
Electricity use (kWh)
300
250
200
150
100
50
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Month
The plot reveals an abrupt change in power usage in June to July. During this period, 153
OUTLIERS
- Individual data points that stand out from the general body
http://support.minitab.com/en-us/minitab-express/1/help-and-how-to/graphs/time-series-
plot/interpret-the-results/key-results/
154
SMOOTHING TIME SERIES
• Most of the time, time series plots look pretty messy, and
• Two methods:
155
MOVING MEAN SMOOTHING
156
MOVING MEAN SMOOTHING
2 13
7 + 13 + 6 + 14 = 40 / 4 = 10
3 6 (10 + 9.88)/2 = 9.94 9.94
13 + 6 + 14 + 6.5 = 39.5 / 4 = 9.88
4 14
5 6.5
157
MOVING MEDIAN SMOOTHING
• Uses the graphical representation of data points to find a
158
MOVIND MEDIAN
2015 VCAA SMOOTHING
Exam 1
160
SEASONAL INDICES
𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑠𝑒𝑎𝑠𝑜𝑛
161
SEASONAL INDICES
Interpreting:
163
SEASONAL INDICES
Correcting:
Warning: People always screw this up, make sure you get it!
164
DESEASONALISATION
165
DESEASONALISATION
- To deseasonalise:
166
MAKING FORECASTS (LINE OF BEST FIT)
167
EXAMPLE
Summer ’11 Autumn ’11 Winter ‘11 Spring ‘11
SI
168
EXAMPLE
Summer ’11 Autumn ’11 Winter ‘11 Spring ‘11
120/1.24 = 96.77
93/0.96 = 96.88
65/0.67 = 97.01
108/1.12 = 96.43
169
EXAMPLE
1 2 3 4
170
EXAMPLE
1 2 3 4
97 – 0.089 × 8 = 96.23
171
EXAMPLE
1 2 3 4
172
plot.
350
Electricity use (kWh)
Seasonal
e present when indices
there are individual and
values thatdeseasonalisation
stand out from the general body Forecasting
173
ies plot below shows the daily power bill for a house (in kWh) for a fortnight.
THAT’S IT!
174
QUESTIONS
GOOD LUCK <3
Presented by:
Josh Hamilton