Unit (2019)
Information on this page, including unit offerings, is from the 2019 academic year.
Time Series Analysis (MAS352)
Organisational Unit  Information Technology, Mathematics and Statistics  
Credit Points  3  
Availability  MURDOCH: S1internal, S1external  
Teaching Timetables  Murdoch S1 

Description  This unit introduces methods for the analysis of measurements made sequentially over time and a brief look at data which are multivariate and correlated. Time series models are developed and inferential procedures (estimation, testing, forecasting) discussed. The unit also includes frequency domain methods for the extraction of signals from noisy data and discussion of linear systems. Applications of time series analysis are shown on Perth temperature data illustrating concepts useful for forecasting and demonstrating aspects of climate change.  
Unit Learning Outcomes  On successful completion of the unit students should be able to: 1. Recognize and understand the nature of correlated data. 2. Recognize autocorrelated data or data which are correlated over time (as is in the nature of many time series data) and learn techniques  of both a simple nature and of a more complex nature, by way of which one may analyse such data. These techniques can include straight forward regression in time or further more appropriate methods in some contexts which involve what is known as the 'Time Domain Approach'. 3. Recognize cyclical patterns that are analysed in the frequency domain. Ideas of spectral analysis and the 'Frequency Domain Approach' are countenanced. 4. Understand bivariate processes that can be analysed using the Cross Spectrum and these lead into the theory of Linear Systems in both the Time Domain and the Frequency Domain. 

Timetabled Learning Activities  Lectures: 3 x 1 hour per week; laboratories/tutorials: 1 x 1 hour per week.  
Unit Learning Experiences  This unit uses a mixture of structured activities and assessments to assist students in learning the material covered in the unit. Structured activities include lectures and tutorials, and assessments include assignments, a project, and a final exam. It is essential that internal students attend lectures and tutorials. External and internal students should review the recorded lectures on Echo360 and study the unit notes and textbook in the progression of the lectures. It is important that students grapple with the computing part of the unit early on so that they may prepare themselves for doing the project later in the semester. By attempting questions at the back of each chapter of the text, students get valuable insight into the material covered in lectures, and ultimately have a good background knowledge for preparing themselves for the final examination.  
Assessment  All students' abilities to correctly apply time series methods will be assessed at regular intervals during the semester via assignments and a project. These assessments are designed to allow students to demonstrate their ability in each of the content areas of the unit and to give them regular feedback on their progress, helping them to identify their areas of strength or weakness during the semester. Assignment solutions and results will be posted progressively on the Learning Management System. The weightings for assessment items are as follows: Assignments (4)  20% Project  10% Final Examination  70% 

Prerequisites  MAS222/MAS278 Probability and Statistical Inference OR MAS223 Applied Statistics OR MAS224/MAS230 Biostatistical Methods OR MAS284 Applied Statistics and Process Management or enrolment in a postgraduate IT course. In addition students must have a calculus background equivalent to at least either MAS161 Calculus and Matrix Algebra OR MAS221/MAS208 Mathematical Modelling.  
Exclusions  Students may not enrol in this unit and either of MAS180 Introduction to Statistics or MAS183 Statistical Data Analysis concurrently. Students who previously successfully completed MAS368 Time Series and Multivariate Analysis cannot enrol in this unit for credit.  
Previously  2015: 'Time Series and Multivariate Analysis'  
Appears in these Courses/Majors: see individual structures for context 


Internet Access Requirements  Murdoch units normally include an online component comprising materials, discussions, lecture recordings and assessment activities. All students, regardless of their location or mode of study, need to have access to and be able to use computing devices with browsing capability and a connection to the Internet via Broadband (Cable, ADSL or Mobile) or Wireless. The Internet connection should be readily available and allow large amounts of data to be streamed or downloaded (approximately 100MB per lecture recording). Students also need to be able to enter into online discussions and submit assignments online. 
Contacts
Unit Coordinator  

No coordinators found for this unit.  
Unit Contacts  
No contacts found for this unit. 