bbricks (version 0.1.4)

dPosteriorPredictive.GaussianNIG: Posterior predictive density function of a "GaussianNIG" object

Description

Generate the the density value of the posterior predictive distribution of the following structure: $$x \sim Gaussian(X beta,sigma^2)$$ $$sigma^2 \sim InvGamma(a,b)$$ $$beta \sim Gaussian(m,sigma^2 V)$$ Where X is a row vector, or a design matrix where each row is an obervation. InvGamma() is the Inverse-Gamma distribution, Gaussian() is the Gaussian distribution. See ?dInvGamma and dGaussian for the definitions of these distribution. The model structure and prior parameters are stored in a "GaussianNIG" object. Posterior predictive density is p(x|m,V,a,b,X).

Usage

# S3 method for GaussianNIG
dPosteriorPredictive(obj, x, X, LOG = TRUE, ...)

Arguments

obj

A "GaussianNIG" object.

x

numeric, must satisfy length(x) = nrow(X).

X

matrix, must satisfy length(x) = nrow(X).

LOG

Return the log density if set to "TRUE".

...

Additional arguments to be passed to other inherited types.

Value

A numeric vector, the posterior predictive density.

References

Banerjee, Sudipto. "Bayesian Linear Model: Gory Details." Downloaded from http://www. biostat. umn. edu/~ph7440 (2008).

See Also

GaussianNIG, dPosteriorPredictive.GaussianNIG, marginalLikelihood.GaussianNIG

Examples

# NOT RUN {
obj <- GaussianNIG(gamma=list(m=0,V=1,a=1,b=1))
X <- 1:20
x <- rnorm(20)+ X*0.3
## out1 and out2 it should have the same values:
out1 <- dPosteriorPredictive(obj = obj, x = x,X=X,LOG = TRUE)
out2 <- numeric(length(x))
for(i in 1:length(x))
out2[i] <- marginalLikelihood(obj,x=x[i],X=X[i],LOG = TRUE)
max(abs(out1-out2))
# }