module specification

The module introduces the students to financial forecasting using modern statistical modelling techniques. Its aim is to prepare the student for work in a quantitative commercial or scientific environment.  Students will be developing problem solving skills.  For each given problem, the process of dealing with it includes, searching for appropriate data sets, establishing the right statistical/financial techniques to use, fitting appropriate models, critically appraising the models using diagnostic model tools and finally interpreting the models and drawing conclusions

MA4057 Data Analysis

Introduction: time series, forecasting and its applications. Time Series Models: Trend, seasonal, cycle and error components; decomposition methods. Smoothing Methods: moving averages, single and double exponential smoothing, Holt-Winters. LO1,LO2, LO3, LO4

Regression Methods: Review of the classical normal, linear regression model and its assumptions. Weighted Least Squares, Autocorrelation and tests such as the Durbin-Watson. Multicollinearity, stochastic regressors. Indicator Variables. LO1,LO2, LO3, LO4

Box-Jenkins methodology; stationarity, autocorrelation, autoregression, ARMA and ARIMA models; model identification and diagnostic checking. LO1,LO2, LO3, LO4

Introduction to state space modelling. LO1,LO2, LO3, LO4

Forecasting Practice: Monitoring forecasts, forecasting accuracy, combining forecasts, comparative study of forecasting methods. LO6

Statistical Packages: use of a statistical package such as, Excel/SPSS, R, to analyse data, fit and validate models and produce forecasts. LO6

The module will be delivered through a weekly 3-hour block consisting of a mixture of a lecture and a tutorial or workshop (in a computer lab). While the theory and methods will be covered during the lectures, the practice exercises and student-led group discussions will be carried out in tutorial and workshop sessions. The tutor-led sessions are intended to enable students to use problem solving skills, whereas the student-led sessions are designed to give students the opportunity for independent learning as well as collaboration. Materials for learning are provided through an integrated learning environment (currently WebLearn) supplemented by online sources and textbooks. In addition to the timetabled classes, students are required to spend about 7 hours each week working individually and in their groups. Then the tutorial sessions and workshops will provide an ideal setting for students to meet up with their group members on regular basis and carry out their discussions and investigations.
The following methods of delivery will be used:
1. Lectures/Contact hours (20 hours). The various methods, together with the case study material, will be discussed each week by the group.
2. Practical Sessions (24 hours) Various data sets will be analysed during the course to demonstrate the forecasting methods. The data sets will cover a range of applications and degrees of complexity. Students will be expected to produce written summaries of their results from the practical sessions.
3. Self-directed study (106 hours). Students will be expected to research a particular self-selected data set and produce a detailed analysis of these data presented in a report.
4.Tutorials (24 hours). Tutorial time will be used to develop students’ awareness of the forecasting process via case study material, self-assessment exercises and discussions of the problems which they encounter during their self-directed study (3 above).

On successful completion of the module, students should be able to
LO1: Demonstrate familiarity with a range of forecasting methods.
LO2:  Select and apply an appropriate method, given a particular problem 
LO3: Critically evaluate the various methods and appreciate their applications and limitations.
LO4: Use data sources and understand their limitations.
LO5: Use a statistical package to analyse time series data, fit and validate models and use the fitted models to forecast.
LO6 Communicate effectively about the subject in general and the results of a specific investigation in particular

Core textbook:
Shumway R and Stoffer D (2010, 3rd Edition). Time series Analysis and its Applications.
https://research.ebsco.com/linkprocessor/plink?id=59a73d50-3cd1-30ee-bf4a-bd500e52d1cf

Recommended reading:
Hamilton, J.D. (2020) Time Series Analysis / James Douglas Hamilton. Princeton University Press. Available at: https://research.ebsco.com/linkprocessor/plink?id=16548d22-8f93-38ba-bc2b-baebbd21d045 (Accessed: 5 December 2025).

Prakash, D.P. (2017) Practical time series analysis : master time series data processing, visualization, and modeling using Python / Dr. Avishek Pal, Dr. PKS Prakash. Packt Publishing. Available at: https://research.ebsco.com/linkprocessor/plink?id=ca8a4d26-3c1b-3709-ac9e-88df1349312a (Accessed: 5 December 2025).
Palma, W. (2016) Time series analysis / Wilfredo Palma. Wiley. Available at: https://research.ebsco.com/linkprocessor/plink?id=0a21bf3b-83d4-3970-bf25-052e6b457744 (Accessed: 5 December 2025).
Cryer, J. (2008) Time Series Analysis with Applications in R.  Springer.   https://www.vlebooks.com/Product/Index/1922083?page=0&startBookmarkId=-1

Makridakis, S, Wheelwright, S.C., Hyndman, R.J. (1998) Forecasting Methods and Applications, 3rd Edition,   Wiley.
Bowerman, B. L., O’Connell, Time Series Forecasting , Duxbury Press, (1987).
Box, G. E. P., Jenkins G. M., Time Series Analysis, Forecasting and Control, Holden Day, (1976).
Cryer, J. Time Series Analysis, Duxbury Press (1986).
Makridakis, S, Wheelwright , S.C., Hyndman, R.J. Forecasting Methods and Applications, 3rd Edition, Wiley, 1998.
Mills, T. C., The Econometric Modelling of Financial Time Series, Cambridge University Press, (1993).
Students will be expected to access materials such as Monthly Digest of Statistics, Annual Abstract of Statistics, Journal of Forecasting, Operational Research Quarterly, etc.
Forecasting Methods and Applications web site: http://www.maths.monash.edu.au/~hyndman/forecasting.