J. Jewson
A challenge when using graphical models in applications is that the sample size is limited relative to the number of parameters. Our motivation stems from applications where one has external data, in the form of networks between variables, that provides valuable information to help improve inference. Specifically, we depict the relation between COVID-19 cases and social and geographical network data, and between stock market returns and economic and policy networks extracted from text data. We propose a graphical LASSO framework where likelihood penalties are guided by the external network data and a spike-and-slab prior framework that depicts how partial correlations depend on the networks, which helps interpret the fitted graphical model. Our applications show how incorporating network data can improve interpretation, statistical accuracy, and out-of-sample prediction, in some instances using significantly sparser graphical models than would have otherwise been estimated
Keywords: GLASSO; Bayesian Inference; Spike-and-Slab
Scheduled
GT04.BIO1 Invited Session. Complex Data
November 7, 2023 3:30 PM
CC3: Room 1