Nonlinear Modeling: Genetic Programming vs. Fast Evolutionary Programming
Minglei Duan, Richard J. Povinelli · e-Publications@Marquette (Marquette University) · 2001
Both Genetic Programming (GP) and Fast Evolutionary Programming (FEP) combined with a Reduced Parameter Bilinear (RPBL) model have been recognized as effective time series modeling methods. This study compares the performance of these two methods for their ability to model time series data in terms of their accuracy and time efficiency. A brief review of GP and FEP are presented. Then the accuracy and time efficiency of these two methods are evaluated on several different time series. The performances of the two methods are compared against each other.