Dynamic Mixture of Experts Models for Online Prediction

A mixture of experts models the conditional density a response variable using finite regression with covariate-dependent weights. We extend model by allowing parameters in both components and weights to evolve time following random walk processes. Inference for time-varying richly parameterized is challenging. propose sequential Monte Carlo algorithm online inference based on tailored proposal distribution built ideas from linear Bayes methods EM algorithm. The method gives unified treatment mixtures essentially any components, including special case static parameters. assess properties simulated data industrial where aim predict software faults continuously upgraded large-scale project.