Abstract:
|
In the era of Big Data, variable selection with high-dimensional data has drawn increasing attention. In this study, we propose a hybrid search algorithm to perform Bayesian high-dimensional variable selection under generalized linear models for various types of outcomes, including binary, count, and continuous data. Using Bayesian approximation techniques, we develop a novel computing strategy that enables us to evaluate all the marginal likelihoods of the neighborhood simultaneously in a single computation. In addition, to accelerate the convergence of our algorithm, we employ a hybrid search algorithm of deterministic local search and stochastic global search. Simulation studies and a real data example are shown to investigate the performance of the newly-developed method.
|