Bayesian Parameter Estimation

Definition: A Bayesian approach to estimating parameter values by updating a prior belief about model parameters (i.e., prior distribution) with new evidence (i.e., observed data) via a likelihood function, resulting in a posterior distribution. The posterior distribution may be summarised in a number of ways including: point estimates (mean/mode/median of a posterior probability distribution), intervals of defined boundaries, and intervals of defined mass (typically referred to as a credible interval). In turn, a posterior distribution may become a prior distribution in a subsequent estimation. A posterior distribution can also be sampled using Monte-Carlo Markov Chain methods which can be used to determine complex model uncertainties (e.g. Foreman-Mackey et al., 2013).

Related terms: Bayes Factor, Bayesian inference, Bayesian statistics, Null Hypothesis Significance Testing (NHST)

References: Foreman-Mackey et al. (2013), McElreath (2020), Press (2007), & https://blog.stata.com/2016/11/15/introduction-to-bayesian-statistics-part-2-mcmc-and-the-metropolis-hastings-algorithm/

Drafted and Reviewed by: Alaa AlDoh, Mahmoud Elsherif, Helena Hartmann, Dominik Kiersz, Meng Liu, Ana Todorovic, Markus Weinmann

Note that we are currently working on an automated mechanism to link references cited above with their full-length version that can be found at https://forrt.org/glossary/references with all references used so far.