Skip to main content
Unit of study_

QBUS2820: Predictive Analytics

Predictive analytics are a set of tools to enable managers to exploit the patterns found in transactional and historical data. For example major retailers invest in predictive analytics to understand, not just consumers' decisions and preferences, but also their personal habits, so as to more efficiently market to them. This unit introduces different techniques of data analysis and modelling that can be applied to traditional and non-traditional problems in a wide range of areas including stock forecasting, fund analysis, asset allocation, equity and fixed income option pricing, consumer products, as well as consumer behaviour modelling (credit, fraud, marketing). The forecasting techniques covered in this unit are useful for preparing individual business forecasts and long-range plans. The unit takes a practical approach with many up-to-date datasets used for demonstration in class and in the assignments.

Code QBUS2820
Academic unit Business Analytics
Credit points 6
QBUS2810 or ECMT2110 or DATA2002
Assumed knowledge:
This unit assumes mathematical knowledge at the level of the Maths in Business program (including calculus and matrix algebra) and basic computer programming skills at the level of QBUS2810

At the completion of this unit, you should be able to:

  • LO1. select and use the appropriate technique to analyse the structure of multivariate data especially when individual data points are identified as belonging to different classes (e.g. failure in credit repayments).
  • LO2. apply multivariate data techniques using a training data set to predict classifications for real data
  • LO3. understand the characteristics of time-series data in order to analyse real business data of this form
  • LO4. select and use an appropriate technique to predict the future behaviour of business variables of interest, including the prediction of discrete outcomes.