INFORMATION CRITERION FOR MODEL SELECTION WITH CONTROLLER DESIGN

摘要

In this paper, an information criterion based on AIC/TIC is proposed which addresses not only on the estimation and selection of model for plant but also on designing of controller. A key point is that the model is estimated on the space of control performance, and the resultant criterion also contains a bias term as AIC/TIC however it depends on the controller. Moreover, when the controller is also designed with stochastic data generated by the estimated model, the criterion is added by another bias term, which is proportional to the dimension of the control parameter.