Zoé van Havre

Zoe van HavrePosition

Joint/Cotutelle PhD student at Queensland University of Technology and Université Paris Dauphine. (E.T.A. Mid-2015)

Supervisors:

  • Prof Kerrie Mengersen,
  • Prof Judith Rousseau,
  • Dr Nicole White.

Qualifications

  • BSc in Statistics, University of Otago
  • BSc (Honour) In Bioinformatics, Griffith University

About

My current interests are in Bayesian methodology and applications relating to the analysis of mixture models and hidden Markov models, particularly when the number of components therein is unknown. My PhD focuses on a comprehensive suite of methods for modelling such situations using overfitted models which are able to deal with the myriad issues arising from overfitting. Key concepts include the mixing ability of MCMC samplers, and dealing with the label switching problem.

Personally, I am driven by one main goal: to simplify the complex and provide tools to improve our understanding of the world around us. I am particularly interested in developing innovative methods which can be used in or adapted for a broad range of applications, supplying tools which are able to be used and understood by those actually employing them in practice. I aim to create useful, logical tools for complex data analysis, which can contribute to the way we process and understand the ever increasing amount of information available around us.

I am experienced with the program R .

Research Interests

  • Development of new methodology
  • Mixture models
  • Hidden Markov Models
  • Bioinformatics and genetics
  • Medical research
  • Biosecurity (environmental protection)

Current Projects 

Most of my work is based on the development of new methods and is theoretical or simulation-based. I am actively working on

  • Overfitting Mixture models
  • Overfitting hidden Markov models
  • Label Switching
  • Parallel Tempering
  • Overfitted VS Non-Parametric mixture models

I am also currently involved in a collaborative project with the CSIRO working on finding subgroups in the controls of an Alzheimer’s research study. This will complement the larger goals of their program, aiming to identify control individuals at risk of having either been misclassified as healthy controls, or who may have developed Alzheimer’s during the course of the study.

Contact Details

zoevanhavre@gmail.com

 

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s