bcgpmodel: Create an object of class bcgpmodel
In cbdavis33/bcgp: Bayesian Composite Gaussian Processes

Description Usage Arguments Details Value See Also Examples

bcgpmodel returns an instance of S4 class bcgpmodel

1	bcgpmodel(x, y, composite = TRUE, stationary = FALSE, noise = FALSE)

`x`	An n x d numeric matrix, where d is the dimension of the data, and n is the number of training data observations.
`y`	A numeric vector of length n.
`composite`	A logical, `TRUE` for a composite of a global process, a local process, and an error process, `FALSE` for non-composite. Defaults to `TRUE`.
`stationary`	A logical, `FALSE` for a non-stationary process, `TRUE` for a stationary process. If `FALSE`, the variance for the process is σ^2(x), and if `TRUE`, the variance is σ^2. Defaults to `FALSE`.
`noise`	If the data should be noise-free (such as from a deterministic computer model), then `noise` should be `FALSE`. Otherwise, it should be `TRUE`. Defaults to `FALSE`

This creates an instance of S4 class bcgpmodel that contains user-provided data, user-provided information about the type of model desired to be fit (composite/non-composite, stationary/non-stationary, deterministic/noisy), default values for the hyperparameters for the prior distributions, information about the prior distributions, randomly-generated starting values for the Markov chains, user-provided information about whether the user desires the data to be scaled before it is fit, and user- provided information on the number of chains desired to be sampled.

An instance of S4 class bcgpmodel containing the data, default values for all the prior parameters, information about the process, and information about the distributions.

bcgpmodel bcgpfit bcgp

Other preprocessing functions: create_parameter_list, update_inits,bcgpmodel-method

1
2
3

simData <- bcgpsims(composite = TRUE, stationary = FALSE, noise = FALSE)
bcgpmodel(x = simData@training$x, y = simData@training$y,
          composite = TRUE, stationary = FALSE, noise = FALSE)