Overview
Topic status: We're looking for students to study this topic.
Data on the circle can occur in various ways such as wind or wave directions, or times of the day when wrapped on to the 24 hour clock. The key aspect is the one of boundary conditions, 0 degrees is the same as 360 degrees. Although there are many models for data on the circle, such as the wrapped normal distribution, there are some issues, such as multimodality of data, time series of circular data, which have not been not dealt with so extensively. A hierarchical Bayesian approach can make modelling of these aspects of multimodality and sequences of data, somewhat more possible.
The project will consider these aspects with a focus on modelling wave buoy data which is collected off the east coast of Australia on a minute by minute basis and gives the direction of waves. Wave direction is important in determining the erosion and deposition of sand on the east coast. The project will review the literature and current methods and software for fitting circular data and investigate the possibilities of fitting multimodal data and time series data.
- Study level
- PhD, Masters
- Supervisors
- QUT
- Organisational unit
Science and Engineering Faculty
- Research area
- Keywords
- statistics
- Contact
-
Please contact the supervisor.
Professor Tony Pettitt
