Name
Abstract | ||
|---|---|---|
The purpose of model validation is to test the goodness of fit of the identified model and to check whether the model is an appropriate representation of the underlying system. Model validation is a final and essential stage in most system identification procedures. Although model validation for nonlinear temporal systems has been extensively studied, model validation for spatiotemporal systems is still an open question. In this paper, correlation based methods, which have been successfully applied in nonlinear temporal systems are extended and enhanced to validate models of spatiotemporal systems. New nonlinear correlation functions are constructed based on the inputs, one-step-ahead predicted outputs and the residuals to test the dependency between the residuals and inputs/outputs. Examples are included to demonstrate the application of the tests. |
Year | DOI | Venue |
|---|---|---|
2007 | 10.1142/S0218127407019986 | INTERNATIONAL JOURNAL OF BIFURCATION AND CHAOS |
Keywords | DocType | Volume |
spatio-temporal systems, model validation | Journal | 17 |
Issue | ISSN | Citations |
12 | 0218-1274 | 0 |
PageRank | References | Authors |
0.34 | 2 | 2 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Y. Pan | 1 | 0 | 0.34 |
| S. A. Billings | 2 | 365 | 60.58 |