Course overview
In this course students will learn the basics of time series analysis for data that violates the assumption of independence through temporal autocorrelation. Commencing with an introduction to temporally autocorrelated data and AutoRegressive Integrated Moving Average (ARIMA) models, content includes extensions to cope with seasonality (SARIMA), vector autoregressive models (VARMA) and conditional heteroskedasticity (ARCH) models. Students will learn the underpinning mathematical frameworks that are used to build these models, will implement the methods in R and produce written summaries of your analysis in the style of professional reports. This supports the program learning outcomes with special emphasis on higher level, technical communication of statistical reports in an unbiased way.
Course learning outcomes
- Evaluate features of a time series using appropriate mathematical formula
- Identify appropriate analysis techniques for data types with differing temporal features
- Effectively analyse non-stationary, complex, time series data using R
- Produce professional statistical reports which evaluate and summarise time series analysis for both specialist and non-specialist audiences