This paper is concerned with the model selection and model averaging problems in system identification and data-driven modelling for nonlinear systems. Given a set of data, the objective of model selection is to evaluate a series of candidate models and determine which one best presents the data. Three commonly used criteria, namely, Akaike information criterion, Bayesian information criterion and an adjustable prediction error sum of squares (APRESS) are investigated and their performance in model selection and model averaging is evaluated via a number of case studies using both simulation and real data. The results show that APRESS produces better models in terms of generalization performance and model complexity.

One method of selecting potential reviewers for papers submitted to the Journal of Geophysical Research Space Physics is to filter the user database within the Geophysical Electronic Manuscript System (GEMS) by areas of expertise. The list of these areas in GEMS can be self selected by users in their profile settings. The Editors have added 18 new entries to this list, an increase of 33% more than the previous 55 entries. All space physicists are strongly encouraged to update their profile settings in GEMS, especially their areas of expertise selections, and details of how to do this are provided.

No abstract