Access this on-demand training session for a practical introduction to time series forecasting with IBM SPSS Statistics. Time series methods are widely used across public and commercial sectors to anticipate demand, plan resources and respond to changing patterns. This session explains the core principles and demonstrates how to build, evaluate and refine forecasts using real examples.

You will learn how to visualise time series data, apply smoothing techniques and define periodicity to reveal underlying structure. The session covers pure time series forecasting, interpreting model fit, handling trend shifts caused by unexpected events and enhancing accuracy by incorporating predictor fields within ARIMA models. You will also see how SPSS Expert Modeler automates model selection and how to generate reliable forecasts for practical use.

Designed for students, researchers and analysts who want to perform time series analysis in SPSS, this session provides an accessible, example-led introduction. A full training pack with data files and outputs is included, allowing you to continue learning at your own pace.

Time series forecasting has a long history and a host of applications in both the public and commercial sectors.  It can be used to:

• Estimate the number of inbound hourly calls to a call centre to more effectively plan staff shift patterns
• Improve inventory management by forecasting the sales of specific goods based on seasonal trends and correlating factors such as weather patterns
• Forecasting monthly tourist/visitor numbers based on previous trends as well as changes in currency values or air travel costs

You will learn:

• The principles of time series forecasting
• Visualising time series
• Smoothing techniques
• Defining periodicity
• Pure time series forecasts
• Interpreting output and model fit
• Dealing with trend shifts due to unexpected events
• Using predictor fields with ARIMA modelling
• Using SPSS Expert Modeler
• Generating forecasts