Abstract
In this article we consider Bayesian inference for partially observed Andersson-Madigan-Perlman (AMP) Gaussian chain graph (CG) models. Such models are of particular interest in applications such as biological networks and financial time series. The model itself features a variety of constraints which make both prior modeling and computational inference challenging. We develop a framework for the aforementioned challenges, using a sequential Monte Carlo (SMC) method for statistical inference. Our approach is illustrated on both simulated data as well as real case studies from university graduation rates and a pharmacokinetics study.
Original language | English (US) |
---|---|
Pages (from-to) | 35-54 |
Number of pages | 20 |
Journal | Foundations of Data Science |
Volume | 2 |
Issue number | 1 |
DOIs | |
State | Published - Mar 2020 |
Bibliographical note
KAUST Repository Item: Exported on 2022-10-25Acknowledgements: We acknowledge the contribution of the study team and participants of the RMP-02/MTN-006 study and thank them for sharing the study data. Partial support for this research came from grant numbers U19 AI060614 and UM1 AI106707 from the U.S. National Institutes of Health. GLR was partially supported by grant P30CA006973 from the U.S. National Cancer Institute.