MAS223 UNIT INFO PDF

Preview

Summary

Unit Information for the course, MAS223...

Description

MAS223 Applied Statistics

Unit Information and Learning Guide

Semester 2, 2021

This information should be read in conjunction with the online learning materials which can be found on your MyUnits page.

Unit coordinator Dr Alethea Rea [email protected] Mathematics and Statistics College of SHEE

© Published by Murdoch University, Perth, Western Australia, July 2021.

This publication is copyright. Except as permitted by the Copyright Act no part of it may in any form or by any electronic, mechanical, photocopying, recording or any other means be reproduced, stored in a retrieval system or be broadcast or transmitted without the prior written permission of the publisher.

2

Murdoch University

Contents Unit information Information about the unit Contact details How to study this unit Resources for the unit Study schedule Assessment

Murdoch University

5 6 7 8 9 10

3

4

Murdoch University

Information about the unit Welcome to: MAS223 Applied Statistics

Unit description Applied Statistics contributes to the general aims of the Mathematics and Statistics Programme, namely to offer a broad view of many branches of mathematics and to achieve professional competence in a number of areas. The unit exposes the students to a variety of applied statistics methods, including select topics from experimental design, survey methods and modelling for continuous and categorical variables. Throughout, advanced statistical software will play an important role in data visualisation and analysis. Students will consider topics through the presentation of real research problems from a number of disciplines, gaining valuable experience in the application of statistics in a variety of contexts, including a project simulating statistical problems commonly encountered in the workplace.

Prerequisites You will need to have completed MAS183 Statistical Data Analysis or equivalent. If you have successfully completed MAS223 Applied Statistics, you may not enrol in this unit for credit. If you are in doubt about your qualifications or background preparation, you should contact the unit coordinator as soon as possible.

Aims of the unit The broad aims of this unit are to introduce a variety of applied statistics techniques that are not covered in other Statistics units but are important for specific experimental designs or modelling approaches.

Learning outcomes for the unit On successful completion of the unit, students should be able to: 1. Carry out a variety of statistical analyses using statistical software. In particular, students should be able to: a. analyse data using linear regression, b. apply the bootstrap for variance estimation, c. assess predictive performance using the bootstrap and cross-validation, d. utilise principal components analysis as a means of dimension reduction, e. use discriminant analysis for classification, and f. be familiar with a variety of other statistical methods.

Murdoch University

5

2. Explain conceptually the various statistical methods covered in the unit, the correct application of these methods, and interpret statistical software output. 3. Write a technical report of findings based on statistical analyses.

Graduate attributes developed in the unit This unit will contribute to the development of the following Graduate Attributes. • Communication • Critical and creative thinking • Social interaction • Independent and lifelong learning • In-depth knowledge of a field of study • Interdisciplinarity This unit was originally written by Dr Ryan Admiraal, 2016 and later revised by Dr Nicola Armstrong.

Contact details Unit Coordinators’ contact details Name: Dr Alethea Rea Email: [email protected] Room: 245.3.026 / Science and Computing 3.026 Phone: +61 8 9360 1392

Tutor/Lecturer contact details Your lecturer will be the unit coordinator, Dr Alethea Rea. Your tutor will be Steve Hogan ([email protected]). Administrative contact details For administrative queries about the operation of this unit, please contact The Academic Support Officer College of SHEE Phone: (08) 9360 6603 If you have any queries about your enrolment in this unit, or if you need information about the University in general, please contact the Student Centre: 1300 687 3624.

6

Murdoch University

How to study this unit The unit structure is presented in the MAS223 Unit Notes. These contain descriptions of the unit content for the first few weeks and can be used to supplement the lecture notes for these introductory topics.

Contact time This unit will consist of some pre-recorded mini-lectures (7-18 minutes) and required reading from the textbook (edition one available for free online). In addition, there will be a workshop, Thursday 10:30am – 12:20pm in room 235.3.034, parts of this workshop will be recorded. Times and locations for tutorials are as follows: Tutorial:

Friday Friday (repeat)

1130 - 1220, 1230 - 1320,

122.2.023 122.2.022

Additionally, in the first three weeks there is one workshop per week dedicated to introducing students to the R statistical programming language for those unfamiliar with R. The time and location for this workshop is as follows: Workshop: Thursday

1330 - 1520,

235.3.034

External students and internal students unable to attend classes: Additional classes may be available on request. Please email Dr Alethea Rea ([email protected]) with suggestions.

Time commitment As this is a 3 credit point unit, we expect you to spend on average 10 hours per week for the total weeks of this teaching period (or 150 hours overall) working on this unit.

Attendance requirements There is no attendance requirement, although students will find that timely viewing of prerecorded material and attendance of workshop and tutorials is vital to performing well in the unit. Students enrolled in external mode are welcome to attend any of the on-campus tutorials if they so desire.

Small group and interactive teaching and learning activities Students are encouraged to form study groups to aid in understanding the material, and they are free to work in groups for tutorials. Both internal and external students are encouraged to use the discussion board on the unit webpage to ask questions and assist each other in better understanding the unit content and statistical software.

Murdoch University

7

Unit changes in response to student feedback We encourage students to provide feedback on how the unit can be improved. Students should feel free to contact the unit coordinators at any point throughout the semester to suggest ways in which the unit might be more effectively taught. Change of Enrolment If your circumstances change and you find that you are unable to complete this unit then it is in your best interests to formally withdraw. To do so you can make changes online through MyInfo. The deadlines are given below. You may wish to discuss your progress with the unit coordinator before you withdraw. Important Deadlines Students should be aware of the implications of different dates of withdrawal from the unit. Before & Including 23 August 24 August to 17 October After 17 October After 23 August

Does not show on academic record. Withdrawn appears on academic record. Fail (withdraws are not permitted). Incurs HECS liability.

Resources for this unit To undertake study in this unit, you will need: Optional Textbook James, G., Witten, D., Hastie, T., and Tibshirani, R. (2018). An Introduction to Statistical Learning with Applications in R, 7th Edition. Springer: New York. ISBN: 9781461471370 An electronic version of this textbook is available for free at: https://static1.squarespace.com/static/5ff2adbe3fe4fe33db902812/t/6062a083acbfe82c7195 b27d/1617076404560/ISLR%2BSeventh%2BPrinting.pdf Online resources The following will be provided electronically during the teaching period: • lecture slides • lecture recordings • tutorial and workshop handouts Computing resources This unit will have a significant computing component. For all assessments other than the exam, computations that are required must be carried out using R (http://cran.r-project.org/), a free and widely used statistical programming language, to carry out a variety of statistical computations and operations. For the exam, students will be permitted to use a calculator for problems requiring computations.

8

Murdoch University

Study schedule The following schedule of topics is subject to variability and/or revision. Thus, the following proposed study schedule should be viewed as a rough guide rather than a prescriptive schedule. Session / Teaching Week

Topic/Project/Study theme

1.

Topic 1: Introduction, point estimators, and sampling distributions

2.

Topic 2: Resampling methods, including the jackknife and bootstrap

3.

Topic 3: Linear regression

4.

Linear regression

5.

Topic 4: Model prediction

6.

Cross-validation

7.

Topic 5: Principal component analysis

8.

Topic 6: Discriminant analysis

9.

Topic 7: Cluster analysis

10.

Topic 8: Design-based inference

11.

Special topic

12.

Revision

Murdoch University

Assessment items

Due

Assignment 1

End of Session 4

Assignment 2

End of Session 7

Assignment 3

End of session 10

Project

End of session 12

9

Assessment Assessment for this unit is conducted in accordance with the Assessment Policy.

Schedule of assessment items You will be assessed on the basis of: Assessment item

Description

Assignment 1 Assignment 2 Assignment 3 Project Examination

Analyses Analyses Analyses Analyses

Value

Due

10% 10% 10% 15% 55%

End of session 4 End of session 7 End of session 10 End of session 12 Assessment period

Assessment details Students must complete three assignments, a project, and a final examination. Details for the final examination are provided below. Assignments will be posted on the unit website as they become available. Instructions for completion are given on the individual assignments. Assignments must be neatly written and set out clearly, concisely, and logically. Be sure to keep copies of all assignments submitted for assessment in the off chance that they are lost.

Examination At the end of the semester a three-hour examination will be held, based on all the material of the unit. The examination will be held in November 2021. Unless otherwise announced during the semester, the exam will be held face-to-face on campus. Those students living within 100km of the Murdoch campus are expected to take their examinations at this campus, and the Examinations Office will advise all others about their examination arrangements. Students must show photographic identification at each exam with a Murdoch University student card, driver's licence, or passport. On-campus students without ID must stay until the end of the exam to have their identification verified. Offcampus students without ID will be recorded on the attendance sheet. The examination will be open book. However, students are recommended to take two A4-sized sheets of paper with them to the exam as a primary reference. Students should also bring a calculator to the examination. Students may inspect their marked examination scripts and discuss the marking with the Unit Coordinator within 14 days of the posting of results (Degree Regulation 43 in http://www.murdoch.edu.au/admin/legsln/regs/bachelor.html#assessment.)

10

Murdoch University

For further information about examinations, refer to http://our.murdoch.edu.au/Student-life/Get-organised/About-exams/

Assignment submission Student assignments must be submitted electronically via the Learning Management System through the provided link on the unit webpage. Assignments should be typed and submitted in one PDF file. (No picture files or Word documents.) Assignments will be scrutinised for plagiarism, copying, and collusion. Although students are free to discuss the assignments with each other, the work presented in assignments must be your own, and we take seriously academic misconduct and do not hesitate to report suspected cases of academic misconduct. Accompanying assignments, it is expected that students will submit a second separate file that contains their R script (.R file). Students should confirm that this file can be run without errors as a stand-alone file prior to submission. Your R script will be examined for copying and/or collusion and should be your own work. It should also be tidied up and clearly commented. Assignments and R scripts must be submitted by specified due dates. Late assignments will not be accepted (i.e. will receive 0 marks). Extensions will be granted only on grounds of illness or other exceptional personal circumstances similar to those applying under Degree Regulation 46. Again, be sure to keep copies of all assignments submitted for assessment in the off chance that they are lost.

Project The project will be an in-depth examination of a topic and includes both a variety of statistical analyses and a written report. Students will be given a choice of several different topics with relevant background information and accompanying data. Details of project topics will be released by the end of Session / Teaching Week 6. Students should plan to work on the project progressively over the last six weeks of the semester.

Murdoch University

11

Determination of the final grade Your final result for the unit will be reported by a letter grade. The minimum percentage scores for overall work and for the supervised and unsupervised assessment components are set out in the following table. Notation

Grade

Percentage Range

HD D C P N DNS

High Distinction Distinction Credit Pass Fail Fail

SA SX

Supplementary Assignment Supplementary Exam

80 – 100 70 – 79 60 – 69 50 – 59 Below 50 Fail, the student failed to participate in assessment components that had a combined weighting of 50% or more of the final mark. 40 – 49* 40 – 49*

*The award of the grade of SA or SX shall be at the discretion of the Unit Coordinator except where clause 11.8 applies.

Contributions for individual assessment components are presented in the “Schedule of assessment items” subsection of this document. See Section 11 in the current Assessment Policy regarding grades.

12

Murdoch University...