Author: Graziani, C.
Title: A Simplified Bayesian Analysis Method for Vaccine Efficacy Cord-id: oalocjde Document date: 2020_12_11
ID: oalocjde
Snippet: We describe a simplified Bayesian analysis of vaccine trial data, in which a reparametrization of the Poisson likelihood leads to a factorization in which the protective vaccine efficacy VES and the nuisance parameter appear in different factors. As a consequence the posterior density acquires a factorized form, and marginalization over the nuisance parameter is trivial. Estimates of VES accordingly become a matter of simple manipulations of one-dimensional posterior probability densities. We de
Document: We describe a simplified Bayesian analysis of vaccine trial data, in which a reparametrization of the Poisson likelihood leads to a factorization in which the protective vaccine efficacy VES and the nuisance parameter appear in different factors. As a consequence the posterior density acquires a factorized form, and marginalization over the nuisance parameter is trivial. Estimates of VES accordingly become a matter of simple manipulations of one-dimensional posterior probability densities. We demonstrate the method using the publically-released final Phase III data from the Pfizer/BioNTech and Moderna mRNA vaccines and for the interim data released for the Sputnik V adenovirus vaccine, for SARS-CoV-2.
Search related documents:
Co phrase search for related documents, hyperlinks ordered by date