bcgpsims-class: An S4 class to represent a simulated model
In cbdavis33/bcgp: Bayesian Composite Gaussian Processes

Description Slots See Also Examples

This class contains training and test data simulated directly from the BCGP model along with some other information about the type of model the data was simulated from (composite/non-composite, stationary/non-stationary, deterministic/noisy). It will most commonly be useful after a call to simulate_from_model() or bcgpsims(). The bcgpsims model can then be plotted or sent to the function bcgp_sampling to fit the model.

training: A list with elements x, an n x d matrix containing the simulated training data locations, and y, a vector of length n containing the simulated training data values. If the composite argument was TRUE and the decomposition argument was TRUE in the call to simulate_from_model, training will also have elements yG, yL, and yE that correspond to the Global, Local, and Error process values at the training data locations.
test: A list with elements x, an nTest x d matrix containing the simulated test data locations, and y, a vector of length nTest containing the simulated test data values. If composite was TRUE and decomposition was TRUE in the call to simulate_from_model, test will also have elements yG, yL, and yE that correspond to the Global, Local, and Error process values at the test data locations. Note: yE in test will always be a zero vector. It is only returned for completeness.
parameters: A list containing the parameters used to simulate from the model.
stationary: A logical indicating whether the model is stationary or not.
composite: A logical indicating whether the model is composite or not.
seed: The seed used by the random number generator to generate the data.