Skip to contents

Generalized g-Prior Distribution for Coefficients in BMA Models

Source: R/beta_priors.R
CCH.Rd

Creates an object representing the CCH mixture of g-priors on coefficients for BAS .

Usage

CCH(alpha, beta, s = 0)

Arguments

alpha

a scalar > 0, recommended alpha=.5 (betaprime) or 1 for CCH. The hyper.g(alpha) is equivalent to CCH(alpha -2, 2, 0). Liang et al recommended values in the range 2 < alpha_h <= 4

beta

a scalar > 0. The value is not updated by the data; beta should be a function of n for consistency under the null model. The hyper-g corresponds to b = 2

s

a scalar, recommended s=0

Value

returns an object of class "prior", with the family and hyperparameters.

Details

Creates a structure used for bas.glm.

See also

IC.prior, bic.prior, bas.glm

Other beta priors: EB.local(), IC.prior(), Jeffreys(), TG(), beta.prime(), g.prior(), hyper.g(), hyper.g.n(), intrinsic(), robust(), tCCH(), testBF.prior()

Author

Merlise A Clyde

Examples

CCH(alpha = .5, beta = 100, s = 0)
#> $family
#> [1] "CCH"
#> 
#> $class
#> [1] "TCCH"
#> 
#> $hyper.parameters
#> $hyper.parameters$alpha
#> [1] 0.5
#> 
#> $hyper.parameters$beta
#> [1] 100
#> 
#> $hyper.parameters$s
#> [1] 0
#> 
#> 
#> attr(,"class")
#> [1] "prior"