A practical introduction to Bayesian analysis to give you the tools to perform your own analysis
- Present the main Bayesian concepts and some theoretical results
- Illustrate Bayesian methodology applied to a wide range of pharmaceutical projects (non-clinical and clinical)
- Show how to move from a « p-value »-based decision-making process to a prediction-based one
- Show the standard Bayesian software (Convergence checks, improve convergence, Model selection: DIC, Missing/Censored/BLQ values, Re-parametrization)
- Implementation using SAS proc MCMC and other procedure with Bayesian capabilities, and using the R environment, with Rstan, brms, MCMCglmm, etc
Pierre Lebrun
Pierre Lebrun is Director Statistics at Pharmalex, which Belgium arm is dedicated to statistical expertise. During more than 14 years, Pierre specialized in quality-by-design aspects related to processes and assays, with a strong emphasis on the use of Bayesian statistics to improve knowledge during the process and assay validation stages. Pierre is also a recognized trainer in statistics for the pharmaceutical industry, including design of experiments, Bayesian statistics, statistical process control, and assay development and validation.
Pierre joined the USP panel in charge of the development of the USP 1220 chapter about a holistic approach to assay validation using the concept of the analytical procedure lifecycle. He also gives lecture on DoE at Liège university (Uliège, Belgium) and collaborates with the research team on new chemometrics and statistics methods for analytical development, validation and PAT. Pierre’s work in applied statistics can be found in 150+ international peer-reviewed papers, book chapters and conference proceedings on these subjects. Pierre Lebrun holds a master’s degree in computer sciences and economy, followed by a master in statistics at the university of Louvain-la-Neuve in Belgium. He completed his PhD in statistics from University of Liège.
