CFE 2019: Start Registration
View Submission - CMStatistics
B1475
Title: To choose or not to choose a prior? That's the question! Authors:  Fatemeh Ghaderinezhad - Gent university (Belgium) [presenting]
Abstract: A challenging question in Bayesian statistics is how choosing the prior can affect the posterior distribution. How can the posteriors derived under different priors be similar as nowadays more and more data are collected? One of the newest instruments to answer this question is Stein's method. This crafty method gives the lower and upper bounds on the Wasserstein distance, at a fixed sample size, between two posteriors resulting from different priors (even improper priors). In addition, this method plays an important role in practice. It often occurs that practitioners hesitate between two proposed priors in a given situation. This methodology then allows them to know how different the two priors actually are, and to decide whether or not it is relevant to consider both priors or just stick to one of them. In particular, when hesitating between a simple, closed-form prior and a much more complicated prior. The applicability of this method is shown by conducting some simulation studies.