doit_marginal: DoIt approximation of the marginal density

In sieste/doit: Bayesian Computation using Design of Experiments-Based Interpolation Technique

Description

Approximate the marginal density of one element of 'theta'.

Usage

doit_marginal(doit, param, theta_eval = NULL)

Arguments

doit	An object of class 'doit', see function 'doit_fit'.
param	Parameter name whose marginal density is calculated. The original design data frame must have a column with that name. If integer, 'param' is interpreted as the column index in the *original* design.
theta_eval	Data frame or matrix of parameter values at which to approximate the marginal distribution. Should have a column named 'param'. If 'NULL' (the default) the original design is used.

Value

A data frame of the provided evaluation points and the corresponding DoIt approximation of the marginal density.

Examples

design   = data.frame(x=rnorm(10), y=rnorm(10))
design$f = with(design, exp(-0.5*(x+y)^2))
fit      = doit_fit(design)
mar_x    = doit_marginal(fit, 'x')

sieste/doit documentation built on May 9, 2019, 4:10 p.m.