Heteroskedastic Gaussian Process Modeling and Design under Replication

[Up] [Top]

Documentation for package ‘hetGP’ version 1.0.1

Help Pages

hetGP-package Package hetGP
allocate_mult Allocation of replicates on existing designs
compareGP Likelihood-based comparison of models
cov_gen Correlation function of selected type, supporting both isotropic and product forms
crit_IMSE Sequential IMSPE criterion
deriv_crit_IMSE Derivative of crit_IMSE
find_reps Data preprocessing
IMSE.search IMSE minimization
IMSE_nsteps_ahead h-IMSE with replication
IMSPE Integrated Mean Square Prediction Error
mleHetGP Gaussian process modeling with heteroskedastic noise
mleHetTP Student-t process modeling with heteroskedastic noise
mleHomGP Gaussian process modeling with homoskedastic noise
mleHomTP Student-T process modeling with homoskedastic noise
predict.hetGP Gaussian process predictions using a heterogeneous noise GP object (of class 'hetGP')
predict.hetTP Student-t process predictions using a heterogeneous noise TP object (of class 'hetTP')
predict.homGP Gaussian process predictions using a homoskedastic noise GP object (of class 'homGP')
predict.homTP Student-t process predictions using a homoskedastic noise GP object (of class 'homGP')
sirEval SIR test problem
sirSimulate SIR test problem
update.hetGP Update '"hetGP"'-class model fit with new observations
update.homGP Fast 'homGP'-update
update_horizon Adapt horizon
Wij Compute double integral of the covariance kernel over a [0,1]^d domain