Communications for Statistical Applications and Methods

Indexed in /covered by CAS, KoreaScience & DOI/Crossref:eISSN 2383-4757   pISSN 2287-7843

Fig. 4.

Download original image
Fig. 4. A plot showing the necessary components for computing the Savage-Dickey density ratio in the independent-samples model. Included are the density of the prior for δ under ℋ1 (i.e., a Cauchy(0,1) distribution, depicted as a dashed line) as well as the logspline density estimate for the posterior of δ (depicted as a solid line). The Bayes factor B01 can be computed as the ratio of the ordinates of δ = 0 under the posterior and prior, respectively.
Communications for Statistical Applications and Methods 2019;26:217-38 https://doi.org/10.29220/CSAM.2019.26.2.217
© 2019 Communications for Statistical Applications and Methods