0342595 20240103193448.320100521235959.9 77949408057 000275920200006 10.1198/TECH.2009.08104 15 s. cav_un_epca*02552010040-170652-66 prediction rolling mills Bayesian Dynamic Averaging cav_un_auth*0237107 Raftery A. E. US cav_un_auth*0101124 Adaptivní systémy Department of Adaptive Systems AS AS Department of Adaptive Systems Kárný Miroslav UTIA-B Ústav teorie informace a automatizace AV ČR, v. v. i. cav_un_auth*0212695 Ettler P. CZ http://library.utia.cas.cz/separaty/2010/AS/karny-0342595.pdf cav_un_auth*0001814 1M0572 GA MŠk cav_un_auth*0261683 7D09008 GA MŠk CZ CEZ:AV0Z10750506 We consider the problem of online prediction when it is uncertain what the best prediction model to use is. We develop a method called Dynamic Model Averaging (DMA) in which a state space model for the parameters of each model is combined with a Markov chain model for the correct model. This allows the "correct" model to vary over time. The state space and Markov chain models are both specied in terms of forgetting, leading to a highly parsimonious representation. As a special case, when the model and parameters do not change, DMA is a recursive implementation of standard Bayesian model averaging, which we call recursive model averaging (RMA). The method is applied to the problem of predicting the output strip thickness for a cold rolling mill, where the output is measured with a time delay. BC 2011 4 A 4a 20231122134017.2 http://hdl.handle.net/11104/0185291 STATISTICSPROBABILITY 1.985 0.206 >10.0 0.00558 1.424 4488 34 1.500 Q1 78.067 75.909 Q1 Q1 2010 Soubory v repozitáři: karny-0342595.pdf 77949408057 SCOPUS 000275920200006 WOS cav_un_epca*0255201 Technometrics 0040-1706 1537-2723 Volume 52 Number 1 2010 52 66