Tutorial AM 4: Oct 22 9:00-12:30
An introduction to extreme value analysis in R
Presenter: Kate Saunders is a lecturer in the Department of Econometrics and Business Statistics at Monash University. She is also a Chief Investigator with the ARC Centre of Excellence in Climate Extremes. Her research focus is on statistical modelling of climate extremes; and understanding how the probability of extreme events might be influenced by natural variability and climate change. Other interests include; statistical post-processing of meteorological forecasts, quality control of meteorological data and how to estimate the risk posed by compound weather events. Kate's research improves our understanding of the probability of extreme climate/weather events and helps us to make informed decisions about natural disaster risk.
Tutorial details
In this tutorial will we will learn the basics of fitting extreme value distributions in R. This will include:
- How to fit a Generalised Extreme Value distribution
- How to fit a Generalised Pareto distribution
- Interpretting model diagnostics
- How to perform model selection
- How to plot return period curves
- Tips and tricks for error handling, and
- Practical data considerations for fitting these distributions
- Use the χ function to explore dependence relationships, and
- How to fit a Bivariate Extreme Value Distribution.
Tutorial materials can be found here.