Answers are grounded in this paper and cite their sources.
Bayesian model selection is a tool for deciding whether the introduction of a new parameter is warranted by the data. I argue that the usual sampling statistic significance tests for a null hypothesis can be misleading, since they do not take into account the information gained through the data, when updating the prior distribution to the posterior. In contrast, Bayesian model selection offers a quantitative implementation of Occam's razor.
No takes yet. Share an insight, caveat, or question.
Roberto Trotta (2007) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: