Scales of Evidence for Model Selection: Fisher versus Jeffreys

54 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Bayes factors based on test statistics (2005, methodology): “As Efron and Gous (2001) and Andrews (1994) pointed out, this is also the case in which the evidence that is contained in Bayes factors and p-values is most easily reconciled.”
Fence methods for mixed model selection (2008, methodology): “Datta and Lahiri [6] discussed a model selection method based on computation of the frequentist’s Bayes factor in choosing between a ﬁxed eﬀects model and a random eﬀects model.”
Large-scale inference (2010, methodology): “Efron and Gous (2001) provide a discussion of Fisher’s interpretive scale for p-values, Table 3.1, and its Bayes-factor competitor, Jeffrey’s scale.”
Size, power and false discovery rates (2007, background): “This threshold requires a much stronger level of evidence against the null hypothesis then in standard one-at-a-time testing, where the critical threshold lies somewhere near 3 [11].”