An overview of skew distributions in cluster analysis
Speaker: Dr Sharon Lee
Affiliation: University of Queensland
Abstract
Real data often does not follow the normal distribution. Motivated by various applications and/or theoretical reasons, the literature on flexible distributions have surged in the past two decades, offering a plethora of alternatives to the normal distribution. Of particular interest is the steam of work on skew distributions. This talk will take you on a tour of some of the most popular (multivariate) non-normal distributions that are capable of handling skewness in the data. We will discuss their characterizations, properties, advantages, and limitations. Simulations will be presented to illustrate their abilities to model distinct types of skewness in the data. To keep the talk brief, we will focus on those distributions that have been adopted for cluster analysis.
About Maths Colloquium
The Mathematics Colloquium is directed at students and academics working in the fields of pure and applied mathematics, and statistics.
We aim to present expository lectures that appeal to our wide audience.
Information for speakers
Information for speakers
Maths colloquia are usually held on Mondays, from 2pm to 3pm, in various locations at St Lucia.
Presentations are 50 minutes, plus five minutes for questions and discussion.
Available facilities include:
- computer
- data projector
- chalkboard or whiteboard
To avoid technical difficulties on the day, please contact us in advance of your presentation to discuss your requirements.