Course Title: Predictive Modelling

Part A: Course Overview

Course Title: Predictive Modelling

Credit Points: 12.00

Terms

Course Code	Campus	Career	School	Learning Mode	Teaching Period(s)
MATH2301	City Campus	Undergraduate	145H Mathematical & Geospatial Sciences	Face-to-Face	Sem 2 2016
MATH2301	City Campus	Undergraduate	171H School of Science	Face-to-Face	Sem 2 2018, Sem 2 2020

Course Coordinator: Dr. Yan Wang

Course Coordinator Phone: +61 3 9925 2381

Course Coordinator Email: yan.wang@rmit.edu.au

Course Coordinator Location: 8.9.34

Pre-requisite Courses and Assumed Knowledge and Capabilities

It is recommended students are familiar with elementary statistics knowledge on sampling distribution, estimation and hypothesis test; and regression modelling on simple linear regression, multiple linear regression and preferable logistic regression. Students should be acquainted with using Microsoft Windows and have some exposure to Windows-based statistics packages such as Minitab, SPSS etc. Previous exposure to a programming language, such as R, Matlab, SAS or Python, is useful but not required.

Course Description

 

With the explosion of “Big Data” problems, statistical learning/machine learning has become a very hot field in many scientific areas as well as marketing, finance, and other business disciplines. People with statistical analytics skills are in high demand.

This course will be divided into two parts: the first part starts with the introduction of statistical learning, classification methods, resampling methods, and linear model selection. The methodologies covered in the first part will be implemented and delivered in R. The second part of the course will cover the methodologies that are commonly used in predictive modelling, including decision tree, logistic regression and neural network. It delivers skills required to assemble analysis flow diagrams using the rich tool set of SAS Enterprise Miner for predictive modelling.

This course is also one of the key subjects in the Joint Certificate Program (JCP) co-sponsored by SAS, one of the leading companies in analytics commercial software.  

 

Please note that if you take this course for a bachelor honours program, your overall mark in this course will be one of the course marks that will be used to calculate the weighted average mark (WAM) that will determine your award level. (This applies to students who commence enrolment in a bachelor honours program from 1 January 2016 onwards. See the WAM information web page for more information.)

Objectives/Learning Outcomes/Capability Development

This course contributes to the Program Learning Outcomes for BP245 Bachelor of Science (Statistics); BP083 Bachelor of Science (Mathematics); and BH119 (Bachelor of Analytics (Honours):

Knowledge and technical competence

• an understanding of appropriate and relevant, fundamental and applied mathematical and statistical knowledge, methodologies and modern computational tools.

Problem-solving

• the ability to bring together and flexibly apply knowledge to characterise, analyse and solve a wide range of problems
• an understanding of the balance between the complexity / accuracy of the mathematical / statistical models used and the timeliness of the delivery of the solution.

Communication

• the ability to effectively communicate both technical and non-technical material in a range of forms (written, electronic, graphic, oral) and to tailor the style and means of communication to different audiences. Of particular interest is the ability to explain technical material, without unnecessary jargon, to lay persons such as the general public or line managers..

Information Literacy

• the ability to locate and use data and information and evaluate its quality with respect to its authority and relevance.

Ethics

• develop the cognitive skills to review critically, analyse, consolidate and synthesise knowledge to identify and provide solutions to complex problems with intellectual independence.

Upon completion of this course you should be able to:

 

1. Understand the fundamental concepts in statistical learning, including bias/variance trade-off, supervised and unsupervised learning, regression and classification, and KNN approach. 
2. Build up linear and logistic regression using R, and carry out linear model selections using subset selection approaches,
3. Conduct exploratory data analysis using SAS Enterprise Miner exploration tools,
4. Build up decision tree and regressions using SAS Enterprise Miner tools,  
5. Select and justify appropriate model assessment criteria and compare performance across different models;
6. Pursue further studies in data analytics and related areas.

Overview of Learning Activities

The course will be delivered through a combination of face-to-face lectures and computer lab practice. While attendance at weekly lectures is beneficial, there is an expectation that you will spend more time out of class on this course, in particular on the practice of the package R and SAS Enterprise Miner. Assessment will be distributed on a regular basis to check your understanding of concepts and to provide additional information. The course is supported by the Canvas management system.

Your will undertake 4 hours per week of face-to-face learning through lecture/lab sessions. Meanwhile it is recommended that an average of 6 hours/week of out-of-class studies is expected on independent study.

Overview of Learning Resources

A list of recommended textbooks for this course is provided on Canvas.

All course materials, including lecture notes, lab exercises, practical exercises, assignments will be posted on Canvas.

The statistical package R, RStudio and SAS Enterprise Miner can be accessed from the school computer labs, as well as through the RMIT MyDesktop system anywhere anytime.   

The subject guide: http://rmit.libguides.com/c.php?g=335968&p=2260895

Overview of Assessment

This course has no hurdle requirements.

 

Assessment Tasks

Assessment Task 1: Lab submissions

Weighting 10%

This assessment task supports CLOs 1, 2, 3 4, 5 & 6

Assessment 2: Tests

Weighting 50%

This assessment task supports CLOs 1, 2, 3, 4, 5 & 6

Assessment 3: Examination

Weighting 20%  

This assessment supports CLOs  1, 2, 3, 4 & 5

Assessment 4: Project

Weighting 20%

This assessment supports CLOs 1, 2, 3, 4, 5 & 6