Extract posterior model probability estimates from BayesMfp objects

Share:

Description

Extract posterior model probability estimates (either normalized estimates or sampling frequencies) from BayesMfp objects.

Usage

1
posteriors(BayesMfpObject, ind = 1)

Arguments

BayesMfpObject

a valid BayesMfp object, containing the models the probabilites of which one wants to estimate

ind

ind = 1 means normalized posteriors, ind = 2 means sampling frequencies

Value

The vector of probability estimates.

Author(s)

Daniel Saban\'es Bov\'e

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
## construct a BayesMfp object
set.seed(19)
x1 <- rnorm (n=15)
x2 <- rbinom (n=15, size=20, prob=0.5) 
x3 <- rexp (n=15)
y <- rt (n=15, df=2)

test <- BayesMfp (y ~ bfp (x1, max = 2) + bfp (x2, max = 2) + uc (x3), nModels = 100,
		  method="exhaustive")

## this works:
posteriors(test)

## this must not work:
## SoDA::muststop(posteriors(test, ind=2))

## only if we do model sampling there are model frequencies:
test2 <- BayesMfp (y ~ bfp (x1, max = 2) + bfp (x2, max = 2) + uc (x3), nModels = 100,
         method="sampling")
posteriors(test2, ind=2)

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.