predict.lmcGP: predict.lmcGP

In berchuck/dependentGP: Dependent Gaussian process using the linear model of coregionalization

Description

Predicts future observations from the lmcGP model.

Usage

## S3 method for class 'lmcGP'
predict(object, NewTimes, type = "response",
  Verbose = TRUE, ...)

Arguments

object	a lmcGP model object for which predictions are desired from.
NewTimes	a numeric vector including desired time(s) points for prediction.
type	a character string indicating the type of predictions to be returned. Either, "response" or "derivative".
Verbose	A logical indicating whether MCMC sampler progress should be printed.
...	other arguments.

Details

predict.lmcGP uses Bayesian krigging to predict the various processes.

Value

predict.lmcGP returns a list for type = "response" and a matrix for type = "derivative".

type = "response"

For type = "response", the list contains the predictive samples for gamma, theta, and the observed data y. Each of these are arrays with three dimensions. The first dimension is the number of time points to be predicted, the second the number of latent dimensions, and finally, the third dimensions is the number of posterior samples specified in the original model.

type = "derivative"

An array of the derivative samples. The array has three dimensions. The first dimension is the number of time points to be predicted, the second the number of latent dimensions, and finally, the third dimensions is the number of posterior samples specified in the original model.

Author(s)

Samuel I. Berchuck

berchuck/dependentGP documentation built on May 14, 2019, 5 a.m.