eplogprob {BAS}  R Documentation 
eplogprob
calculates approximate marginal posterior inclusion
probabilities from pvalues computed from a linear model using a lower bound
approximation to Bayes factors. Used to obtain initial inclusion
probabilities for sampling using Bayesian Adaptive Sampling bas.lm
eplogprob(lm.obj, thresh = 0.5, max = 0.99, int = TRUE)
lm.obj 
a linear model object 
thresh 
the value of the inclusion probability when if the pvalue > 1/exp(1), where the lower bound approximation is not valid. 
max 
maximum value of the inclusion probability; used for the

int 
If the Intercept is included in the linear model, set the marginal inclusion probability corresponding to the intercept to 1 
Sellke, Bayarri and Berger (2001) provide a simple calibration of pvalues
BF(p) = e p log(p)
which provide a lower bound to a Bayes factor for comparing H0: beta = 0 versus H1: beta not equal to 0, when the pvalue p is less than 1/e. Using equal prior odds on the hypotheses H0 and H1, the approximate marginal posterior inclusion probability
p(beta != 0  data ) = 1/(1 + BF(p))
When p > 1/e, we set the marginal inclusion probability to 0.5 or the value
given by thresh
.
eplogprob
returns a vector of marginal posterior inclusion
probabilities for each of the variables in the linear model. If int = TRUE,
then the inclusion probability for the intercept is set to 1. If the model
is not full rank, variables that are linearly dependent base on the QR
factorization will have NA for their pvalues. In bas.lm, where the
probabilities are used for sampling, the inclusion probability is set to 0.
Merlise Clyde clyde@stat.duke.edu
Sellke, Thomas, Bayarri, M. J., and Berger, James O. (2001), “Calibration of pvalues for testing precise null hypotheses”, The American Statistician, 55, 6271.
library(MASS)
data(UScrime)
UScrime[,2] = log(UScrime[,2])
eplogprob(lm(y ~ ., data=UScrime))