Skip to content

Latest commit

 

History

History
19 lines (10 loc) · 770 Bytes

README.md

File metadata and controls

19 lines (10 loc) · 770 Bytes

LoALDIG

Code for the proposed LoALDIG (Local Approximate Likelihood Data Integration) method in the paper entitled "Structural inference of time-varying mixed graphical models" authored by Qingyang Liu, Yuping Zhang and Zhengqing Ouyang.

License: GPL-2.

Correspondence: Yuping Zhang, Email: [email protected].

local_tvgm.m: get the local graph structure at a give time point.

refit.m: given a graph structure, refit model to reduce over-shrinkage.

cv.m: cross-validation to select hyperparameters.

sampling.m: simulate synthetic data through Gibbs sampling.

example.m: an example to demonstrate the usage of our code.

Reference: Liu, Q., Zhang, Y. and Ouyang, Z., 2021. Structural inference of time‐varying mixed graphical models. Stat, 10(1), p.e414.