QBUS2820: Predictive Analytics
Applied predictive models and forecasting techniques to business and financial data, focusing on classification, time-series forecasting, and decision-making under uncertainty.
Learning Outcomes
- Multivariate Analysis: Selected and applied techniques to analyse structured data, especially classification problems (e.g. credit default, fraud detection).
- Predictive Modelling: Built models with training datasets to classify and predict real-world business outcomes.
- Time-Series Forecasting: Understood and applied methods for analysing and forecasting business time-series data.
- Business Applications: Predicted discrete outcomes and future behaviour of variables for finance, retail, and consumer analytics contexts.
- Practical Implementation: Worked with up-to-date datasets to translate theory into applied predictive tasks.
Takeaways
Since I am only eligible to officially enrol in this unit next year, I chose to audit it this semester to gain early exposure. Although I was not formally assessed, attending the lectures gave me a meaningful first look at predictive analytics and its broad range of applications in business, from modelling consumer credit risk to forecasting financial returns and marketing behaviour. The lecturer was exceptionally patient and approachable, often answering my questions in detail and clarifying concepts beyond the lecture slides, which helped me build confidence with material that would otherwise have felt overwhelming. The pace of the unit challenged me to keep up, but it also motivated me to consolidate my understanding of multivariate techniques and time-series forecasting. Overall, auditing this course not only broadened my technical awareness but also reinforced my interest in predictive modelling, and I look forward to revisiting the content in depth when I officially enrol next year.