Determining conditional Gaussian distributions for decomposable graphical models: - A new method
Translated title
Determining conditional Gaussian distributions for decomposable graphical models: - A new method
Author
Term
4. term
Education
Publication year
2016
Submitted on
2016-01-11
Pages
37
Abstract
This report introduces a new method for calculating conditional Gaussian distributions for decomposable models. Basic graph theory is explained followed by multivariate normal distribution and maximum likelihood estimation. The new method is then presented along with the general case. Lastly, the implementation in R is explained, and comparison with a simple method is made.
Documents