Mathematics Learning Centre
Support in learning maths and stats
Carslaw building room 455
Drop-in Centre Carslaw room 455
Opening hours
Monday 10am to 5pm
Tuesday 10am to 7pm (5pm in Weeks 1 and 2)
Wednesday 10am to 5pm
Thursday 10am to 5pm
Friday 10

QBUS 6810
Bayesian Learning and EM-Algorithm
Professor Junbin Gao
The University of Sydney Business School
13 October 2016
1 / 34
Outlines
Readings: This slide
Bayesian Linear Regression
Likelihood
Mixture of Gaussians (MoG)
EM-algorithm
2 / 34
Linear Reg

QBUS 6810
Subset Selection and Shrinkage Methods
Professor Junbin Gao
The University of Sydney Business School
18 August 2016
1 / 38
Message for the Final Exam
For the format of final exam papers, we may have
Multiple Choice Questions
Other types of quest

QBUS 6810
Overview of Supervised Learning
Professor Junbin Gao
The University of Sydney Business School
1 / 34
Outlines
Linear Algebra: Matrix
Supervised Learning Setting
Two simple examples: Regression and Classification
Statistical Decision Theory
Curse

QBUS 6810
Overview of Supervised Learning and
Linear Methods for Regression
Professor Junbin Gao
The University of Sydney Business School
11 August 2016
1 / 40
Outlines
Reading: Sections 2.4-2.6, 2.9 and 3.1 - 3.2
OLS vs k-NN
Statistical Decision Theory
C

QBUS 6810
Classification (II)
Professor Junbin Gao
The University of Sydney Business School
1st September 2016
1/1
Outlines
Readings: Section 4.4 and this slide
Logistic Regression for Classification
Probit Regression
Optimal Hyperplane and Support Vector

QBUS6840: Tutorial 4 Linear Regression (1)
Objectives
Apply linear regression models
Analyze and evaluate suitability of linear regression models
Develop Python skills
1. Load the Data and Visual Inspection
Create a new Python script called tutorial_04.py

QBUS6840: Tutorial 2 Data Manipulation and
Plotting
Objectives
Learn concepts of data manipulation
Learn concepts of plotting and visualization
Learn how to use numpy, pandas and matplotlib packages
While Python provides a lot of general functionality it

QBUS6840: Tutorial 3 Time Series Decomposition
Objectives
Learn how to decompose time series data into trend, seasonal and cycle
components
This tutorial follows slides 36 and 37 (Algorithm for Additive Decomposition)
of Lecture 03. It is an extension of

QBUS6840
Semester 1, 2017
Mid-Semester Examination Practice Questions:
Note: The number of questions here is more than the number of questions in the mid-semester
exam paper. The purpose is to give you more questions to practise
1. The file beer.txt conta

QBUS6840: Tutorial 1 - Getting Started
Objectives
To configure and become familiar with Anaconda and Spyder;
To learn fundamental Python programming concepts;
1. Launch Spyder
First open Anaconda Navigator then click the Launch button for Spyder
Spyder wi

QBUS6840 Lecture 6
Exponential Smoothing (Seasonal)
Professor Junbin Gao
The University of Sydney Business School
1 / 44
Outline
Exponential smoothing
Holt-Winters smoothing
Exponential smoothing methods for seasonal data.
Additive seasonality.
Multiplica

QBUS6840 Lecture 03
Decomposition Methods
Professor Junbin Gao
The University of Sydney Business School
1 / 41
Outlines
Drift Method
Time series decomposition: Additive and multiplicative models
Forecasting using decompositions
X11 decomposition
Readings:

QBUS 6840 Lecture 8
ARIMA models (II)
Professor Junbin Gao
The University of Sydney Business School
1 / 59
Sample Questions
Q5(i) A small number of large forecast errors needs to be
guarded against.
From Lecture 2:
MAD: does not heavily penalise a small n

QBUS6840 Lecture 7
ARIMA Models (I)
Professor Junbin Gao
The University of Sydney Business School
1 / 53
Reschedule
Week 7: Stationarity and AR Process: ARIMA Part I
Week 8: MA Process and ARIMA: ARIMA Part II
Week 9: Seasonal ARIMA
Week 10: Forecast Comb

QBUS 6840 Lecture 5
Exponential Smoothing
Professor Junbin Gao
The University of Sydney Business School
1 / 42
Outline
Exponential smoothing
Simple Exponential Smoothing
Trend Corrected Exponential Smoothing (Holts Linear Trend
Method)
Reading
Online Text

QBUS6840 Lecture 4
Time Series Regression
Professor Junbin Gao
The University of Sydney Business School
1 / 67
Lecture 4
Outline:
Simple linear regression.
Issues in time series regression.
Multiple regression.
Some useful time series predictors.
Selectin

QBUS6840 Lecture 02
Data Pattern, Graphing, Time Series Components,
and Forecast Accuracy
Professor Junbin Gao
The University of Sydney Business School
1 / 67
Outlines
Data and Data Pattern
Graphing Data
Components of Time Series
Nave Methods
Moving Avera

QBUS 6810
Basis Expansions and Regularisation
Professor Junbin Gao
The University of Sydney Business School
8 September 2016
1 / 38
Outlines
Readings: Up to Section 5.4 of Chapter 5
Piecewise Polynomials
Splines
Natural Cubic Splines and Natural Boundary

QBUS 6810
Association Rules Mining
Professor Junbin Gao
The University of Sydney Business School
20 October 2016
1 / 37
Outlines
Readings: Sec 14.2 of Our Textbook and Section 6.1-6.3 &
6.7 of Introduction to Data Mining, available at https:
/www-users.cs

The University of Sydney
School of Mathematics and Statistics
Tutorial for Week 3
MATH1011: Applications of Calculus
Semester 1, 2015
Web Page: http:/www.maths.usyd.edu.au/u/UG/JM/MATH1011/
Lecturer: George Papadopoulos, Sharon Stephen
Questions to comple

The University of Sydney
School of Mathematics and Statistics
Solutions to Tutorial for Week 3
MATH1011: Applications of Calculus
Semester 1, 2014
Web Page: http:/www.maths.usyd.edu.au/u/UG/JM/MATH1011/
Lecturer: Clinton Boys, Ross Ogilvie and George Papa

The University of Sydney
School of Mathematics and Statistics
Solutions to Tutorial for Week 9
MATH1011: Applications of Calculus
Semester 1, 2015
Web Page: http:/www.maths.usyd.edu.au/u/UG/JM/MATH1011/
Lecturer: George Papadopoulos, Sharon Stephen
Questi

The University of Sydney
School of Mathematics and Statistics
Solutions to Tutorial for Week 6
MATH1011: Applications of Calculus
Semester 1, 2014
Web Page: http:/www.maths.usyd.edu.au/u/UG/JM/MATH1011/
Lecturer: Clinton Boys, Ross Ogilvie and George Papa

MATH1011 Applications of Calculus
Lecture 1
Introduction and sinusoidal functions
(Original notes by Dr. Emma Carberry)
March 5, 2015
1 / 42
Introduction
Lecturers:
Dr. Sharon Stephen, [email protected]
Mr. George Papadopoulos,
george.papadopou

The University of Sydney
School of Mathematics and Statistics
Solutions to Tutorial for Week 11
MATH1011: Applications of Calculus
Semester 1, 2015
Web Page: http:/www.maths.usyd.edu.au/u/UG/JM/MATH1011/
Lecturer: George Papadopoulos, Sharon Stephen
Revis

The University of Sydney
School of Mathematics and Statistics
Solutions to Tutorial for Week 5
MATH1011: Applications of Calculus
Semester 1, 2014
Web Page: http:/www.maths.usyd.edu.au/u/UG/JM/MATH1011/
Lecturer: Clinton Boys, Ross Ogilvie and George Papa

The University of Sydney
School of Mathematics and Statistics
Solutions to Tutorial for Week 10
MATH1011: Applications of Calculus
Semester 1, 2015
Web Page: http:/www.maths.usyd.edu.au/u/UG/JM/MATH1011/
Lecturer: George Papadopoulos, Sharon Stephen
Quest

QBUS 6810
Introduction and Linear Algebra
Professor Junbin Gao
The University of Sydney Business School
1 / 54
Outlines
Unit House Keeping
Introduction to Data Mining
Linear Algebra Review
2 / 54
Introduction to QBUS6810
3 / 54
Unit Objectives
know the st

The University of Sydney
School of Mathematics and Statistics
Mid-semester-break revision questions
MATH1011: Applications of Calculus
Semester 1, 2015
Lecturer: George Papadopoulos, Sharon Stephen
1. Find the equation of the line that
(a) has gradient 3

QBUS6840 Lecture 01
Introduction
Professor Junbin Gao
The University of Sydney Business School
1 / 58
Outlines
Unit House Keeping
Introduction to Forecasting
Prediction vs Forecasting
Time Series vs Other Data
Qualitative vs Quantitative
Basic concepts of