Forecasting techniques are a key part of decision making, affecting planning at all levels: strategic, tactical and operational. Effective and accurate forecasting techniques are particularly important in a dynamic, changing and uncertain environment. They are indispensable for coordination, execution, and risk management. An understanding of the fundamentals of forecast assessment is vital for all managers and executives. Knowing how to calculate and build different kinds of forecasts is a related, specialised skill.
Examining “what has happened” is a necessary precursor to predicting “what will happen” in the future. Time series analysis is thus an important forecasting foundation.
This course provides a grounding in both time series analysis and forecasting.
The forecasting component of will include the fundamentals of assessing forecast effectiveness. The merits of various error measures will be addressed, along with what they say about the effectiveness of future forecasts, “rolling” forecasts, and forecasts at multiple horizons. A range of forecasting tools will be introduced, including:
- Least squares trend extrapolation
- Naive forecasting
- Moving averages
- Period-on-period ratios
Exponential smoothing will be introduced as an advanced, accurate and highly automated forecasting tool.
The course will additionally provide a range of time series analysis techniques essential for management reporting, planning and monitoring, including:
- Basic decomposition
- Dashboarding techniques
- Trend, seasonality, and variability separation
- Detrending and deseasonalisation
- Visualisation methods supporting time series analysis
Finally, collective forecasting techniques will be introduced, showcasing the way in which forecasts of qualitative events can be produced by pooling the judgement of individuals, and how these can be used to conduct “what if” analyses in the context of strategic decisions.
The course will be conducted using the readily available tools of MS Excel and R, along with Presciient’s System-II collective forecasting platform.
Who should attend?
This course is suitable for all managers, executives and specialists who want to make better decisions under uncertainty. It requires no specialised statistical knowledge, nor knowledge of R.
Attendees will, by the end of the course:
- Understand the fundamentals of time series composition and analysis.
- Know how to assess any forecast, and accordingly how to manage any forecasting team, model, method, or process.
- Understand the strengths, weaknesses, and applications of a range of statistical and judgement-based forecasting techniques.
The course will be instructed by Presciient Director, Dr Eugene Dubossarsky. Eugene is a founder and Fellow of the Institute of Analytics Professionals of Australia (IAPA); Director, University of New South Wales School of Mathematics and Statistics Industry Advisory Board; and a recognised industry leader in Business Analytics. Eugene is an experienced, professional data miner of 20 years’ experience programming in R and its parent language, S.
Attendees are recommended to have completed Presciient’s Introduction to R two-day course, or equivalent.