This package implements dynamic Bayesian model averaging as described for continuous outcomes in Raftery et al. (2010, Technometrics) and for binary outcomes in McCormick et al. (2011, Biometrics).
Tyler H. McCormick, Adrian Raftery, David Madigan, Sevvandi Kandanaarachchi [ctb], Hana Sevcikova [ctb]
Maintainer: Hana Sevcikova <hanas@uw.edu>
The main averaging functions are:
dmafor continuous outcomes
logistic.dmafor binary outcomes
McCormick, T.M., Raftery, A.E., Madigan, D. and Burd, R.S. (2011) "Dynamic Logistic Regression and Dynamic Model Averaging for Binary Classification." Biometrics, 66:1162-1173.
Raftery, A.E., Karny, M., and Ettler, P. (2010). Online Prediction Under Model Uncertainty Via Dynamic Model Averaging: Application to a Cold Rolling Mill. Technometrics 52:52-66.