Bayes Factor for Regression

Sample Size:

Number of Covariates:

Unadjusted R-squared :


Summary: This calculator computes Bayes factor for linear models from Liang et al (2008).

Models:
Null Model: y =1μ + ε
Alternative. Model: y =1μ + Xβ + ε
where,
y: a column vector of N observables
1: a column vector of 1.0, length N
μ: intercept
ε : a column vector of normal iid residuals of length N
X: a NxP design matrix
β: a vector of parameteres of length p.

Priors Jeffrey-Zellner-Siow mixture of g-priors. See Eqs (3), (4), and (15) in Liang et al.

Citation/Details: Liang et al. (2008), Mixtures of g Priors for Bayesian Variable Selection. Journal of the American Statistical Association 103, p. 410-423.

Questions: rouderj@missouri.edu

Last Update 6/10.