Dr Peter Baker

Honorary Senior Lecturer

School of Public Health
Faculty of Medicine

Overview

Dr Peter Baker is an Honorary Senior Lecturer at the School of Public Health, University of Queensland.

For twelve years until the end of 2020, Peter was a Senior Lecturer in Biostatistics at the School of Public Health and a senior statistical collaborator, advisor and consultant to several research projects in the Faculty of Medicine, School of Public Health.

With fourty years experience as a statistical consultant and researcher, Peter has a passion for biostatistics applied to public health and medical research. He also champions reproducible research and reporting and to this end has developed R and Make software to aid the workflow of data analysts in any field. As a statistical consultant and collaborator, he has contributed to many research projects. His contribution has ranged from advice on standard statistical approaches to the application of novel methods to improve statistical analysis or the development of new statistical methodology to fill a gap in the knowledge.

Peter's current research interests:

  • efficient statistical computing using R, Make, Git and related software for the workflow of data analysis,
  • reproducible research and reporting using R, markdown and sweave,
  • tailoring R functions and developing bespoke packages for specific statistical analyses, and
  • applied statistlcal research in novel methods for epidemiological and medial research, including
    • graphical models for multivariate data in epidemiology,
    • statistical methods for modelling trajectories of alcohol consumption in youths,
    • propensity score analysis to adjust for selection bias in observational studies, and
    • Bayesian methods for epidemiological and medical MCMC studies.

Dr Baker is an Accredited Statistician (ASTAT) with the Statistical Society of Australia (see SSAI_Accreditation)

Research Impacts

Peter is a highly experienced statistical consultant and researcher who worked at CSIRO Mathematics, Informatics and Statistics for 20 years before joining the School of Population Health at the University of Queensland. During his time at CSIRO, he worked in a range of areas including agricultural research, bioinformatics and quantitative genetics. Highlights: developed novel Bayesian methods for assessing marker dosage in sugarcane and other polyploids and applied false discovery rate methods to QTL analysis and microarrays.

In addition to collaborating with a range of applied scientists to advance scientific knowledge by employing standard and cutting-edge statistical methods, several projects have had practical significance to industry including timely alerts to graziers when drops in temperature are dangerous for lambing, improved soil moisture estimation for cotton production and enhancing quantitative methods for sugarcane breeding. Recent collaborative work has informed public policy on drink safe precincts and alcohol related problems, effectiveness of mosquito nets and the effect of sunscreen to prevent skin ageing.

When it can improve research outcomes, Peter is keen to apply new statistical methods to epidemiology, health and medical studies in order to extract the maximum information from study data. He is also an expert in, and advocate for, reproducible research and efficient statistical computing using the R statistical and graphics system.

R Clinics and Short Courses

  • Please contact Peter if you would like a workshop or Introductory to Advanced R Course tailored to your needs.

Student Projects

  • Developing or contributing to R packages for various aspects of the workflow of data analysis projects like data wrangling, cleaning, validation and reproducible reporting,
  • Graphical Models for epidemiological and medical studies.

Qualifications

  • Master of Statistics, University of New South Wales
  • Bachelor of Science, The University of Adelaide
  • Doctor of Philosophy, Queensland University of Technology

Publications

View all Publications

Supervision

View all Supervision

Publications

Book

Book Chapter

  • Baker, P., Mengersen, K. and Davis, G. (1999). On a Bayesian method for QTL detection within selectively genotyped half sib families.. Inaugural Australian Workshop on the Application of Artificial Intelligence, Optimisation and Bayesian Methods in Agriculture. (pp. 55-72) edited by Hussein A. Abbass and Michael W. Towsey. Brisbane, QLD, Australia: QUT Publications.

Journal Article

Conference Publication

PhD and MPhil Supervision

Current Supervision

Completed Supervision