predictive.struct {FVDDPpkg}

Use the predictive structure of the FVDDP to sequentially draw values adn update

Description

Use the predictive structure of the FVDDP to sequentially draw values adn update

Usage

predictive.struct(fvddp, N)

Arguments

fvddp

The instance of class fvddp the values are drawn from.

N

The amount of values to draw.

Value

A vector of length N of values obtained using the predictive structure. Precisely, after that any observation is drawn (either from the centering measure or from past observations) the input fvddp is modified as if the function update() is called, with the new value as second argument.

References

Ascolani F, Lijoi A, Ruggiero M (2021). “Predictive inference with Fleming–Viot-driven dependent Dirichlet processes.” Bayesian Analysis, 16(2), 371 – 395. doi:10.1214/20-BA1206.

Examples

#create a dumy process and expoit the predictive structure
FVDDP = initialize(7, function(x) rbeta(x, 3,3),
                   function(x) dgamma(x, 3,3), FALSE)
FVDDP = update(FVDDP, rep(0:1, 2))
predictive.struct(fvddp = FVDDP, N = 100)