MVNBayesian-package

tools:::Rd_package_title("MVNBayesian")

tools:::Rd_package_description("MVNBayesian") package

Details

This package is aimed to build a easy approach for MVN (mixture) distribution in Bayesian analysis framework. Bayesian posteriori MVN (mixture) distribution can be calculated in conditions of given priori MVN (mixture) informations. The conjugated property of MVN distribution makes it effective in parameter estimation using Bayesian iterator. Joint and marginal probability densities of a certain MVN (mixture) can be achieved through random vector generator, using Gibbs sampling. Conditional probability densities from a certain MVN (mixture) can be simulated using MCMC method.

Author(s)

tools:::Rd_package_author("MVNBayesian")

Maintainer: tools:::Rd_package_maintainer("MVNBayesian")

References

"Statistical Inference" by George Casella. Roger L. Berger;

"Statistical Decision Theory and Bayesian Analysis" by James O. Berger;

"Matrix Computation" by Gee H. Golub. Charles F. Van Loan;

"Bayesian Statistics" by WEI Laisheng;

"Machine Learning" by NAKAGAWA Hiroshi.

See Also

stats, mvtnorm

Examples

library(Rfast)
library(mvtnorm)
library(plyr)

head(dataset1)

BP <- MVN_BayesianPosteriori(dataset1)
BP

BP_Gibbs <- MVN_GibbsSampler(5000, BP)
colMeans(BP_Gibbs)
colrange(BP_Gibbs)

result <- MVN_MCMC(BP, 5000, c(1), c(77.03))
result$Accept