A Bayesian heteroscedastic composite Gaussian process is introduced that combines global and local GPs with a variance process for non-stationary prediction and MCMC-based posterior inference.
O., Gelman, A., and Gilks, W
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ME 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Prediction Using a Bayesian Heteroscedastic Composite Gaussian Process
A Bayesian heteroscedastic composite Gaussian process is introduced that combines global and local GPs with a variance process for non-stationary prediction and MCMC-based posterior inference.