Description
Bayesian Spectral Analysis Models using Gaussian Process Priors.
Description
Contains functions to perform Bayesian inference using a spectral analysis of Gaussian process priors. Gaussian processes are represented with a Fourier series based on cosine basis functions. Currently the package includes parametric linear models, partial linear additive models with/without shape restrictions, generalized linear additive models with/without shape restrictions, and density estimation model. To maximize computational efficiency, the actual Markov chain Monte Carlo sampling for each model is done using codes written in FORTRAN 90. This software has been developed using funding supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (no. NRF-2016R1D1A1B03932178 and no. NRF-2017R1D1A3B03035235).
README.md
bsamGP README
Authors
* Seongil Jo <[email protected]>
* Taeryon Choi <[email protected]>
* Beomjo Park <[email protected]>
* Peter J. Lenk <[email protected]>
Acknowledgments
* Research of Seongil Jo was supported by Basic Science Research Program
through the National Research Foundation of Korea (NRF) funded by
the Ministry of Education (NRF-2017R1D1A3B03035235)
* Research of Taeryon Choi was supported by Basic Science Research Program
through the National Research Foundation of Korea (NRF) funded by
the Ministry of Education (NRF-2016R1D1A1B03932178)
Changelog
* v1.0.1 -- 1.0.2
* Add missing dependencies to src/Makevars. (Thanks to Prof. Brian Ripley)
* v1.1.0
* Change input UI to R formula.
* Add S3methods : predict, summary
* Clean plot method.
* v1.1.1 -- 1.1.3
* Separate predictbsam & predictgbsam : BUGFIX
* Add verbose option.
* Minor Bug fix
* v.1.2.0 -- 1.2.4
* Supports Multiple extreme shapes for bsar() and bsaq()
* Supports scalable regression function bsarBig()
* Memory bug fix
* Minor bug fix
* gcc/gfortran 12 compatibility check