Intro to Computational Bayesian Methods for Chemistry, Manufacturing & Controls_Part 4 of 4 скачать в хорошем качестве

Intro to Computational Bayesian Methods for Chemistry, Manufacturing & Controls_Part 4 of 4

Webinar presented on August 6, 2021. This webinar is intended for statisticians who are interested in Bayesian computation and who want to learn how to conduct Bayesian analysis in the area of nonclinical statistics. The first webinar examines some questions of interest in the nonclinical framework and reviews key differences between Bayesian and frequentist treatments of these issues. In addition, software plays an important role in the modern-day Bayesian paradigm, and we devote the second part of the first webinar to introduce some useful computational tools. PROC MCMC is a general-purpose procedure that is well suited to simulating and exploring Bayesian models. PROC BGLIMM provides full Bayesian inference for generalized linear mixed-effects models. We introduce these procedures and describe how to use them for estimation, inference, and prediction. The second and third webinars use a topic-driven approach to explore several case studies in the biotech and pharmaceutical field, focusing on nonclinical applications. Topics include setting acceptance criteria on the basis of different distributions and small sample sizes, estimating prevalence rates, making posterior predictions that a product lot is within specifications, using historical information in analyses, and fitting stability models. The main objective of this webinar is to familiarize attendees with the essentials of Bayesian techniques and to equip them with useful Bayesian computational tools through worked-out examples drawn from common situations in the CMC area of the pharmaceutical industry. IQ Consortium Statistics Leadership Group BIO & CMC Fora Learn more at iqconsortium.org