BKP: Beta and Dirichlet Kernel Processes for Binomial and Multinomial
Modeling
Provides methods for nonparametric modeling of binomial and multinomial success probabilities via the Beta Kernel Process and its extension, the Dirichlet Kernel Process.
Supports model fitting, predictive inference with uncertainty quantification, posterior simulation, and visualization in one- and two-dimensional input spaces.
The package implements multiple kernel functions (Gaussian, Matern 5/2, and Matern 3/2), and performs hyperparameter optimization using multi-start gradient-based search.
Applications include spatial statistics, probabilistic classification, and Bayesian experimental design.
For more details, see MacKenzie, Trafalis, and Barker (2014) <doi:10.1002/sam.11241>.
Documentation:
Downloads:
Linking:
Please use the canonical form
https://CRAN.R-project.org/package=BKP
to link to this page.