Abstract
A new open box and nonlinear model of cosine and sigmoid higher order neural network (CS-HONN) is presented in this chapter. A new learning algorithm for CS-HONN is also developed in this chapter. In addition, a time series data simulation and analysis system, CS-HONN simulator, is built based on the CS-HONN models. Test results show that the average error of CS-HONN models are from 2.3436% to 4.6857%, and the average error of polynomial higher order neural network (PHONN), trigonometric higher order neural network (THONN), and sigmoid polynomial higher order neural network (SPHONN) models range from 2.8128% to 4.9077%. This suggests that CS-HONN models are 0.1174% to 0.4917% better than PHONN, THONN, and SPHONN models.
Reference27 articles.
1. Polynomial and neural networks: analogies and engineering applications.;R.Barron;Proceedings of International Conference of Neural Networks,1987
2. Approximation theory and feedforward networks
3. Dynamic ridge polynomial higher order neural network;R.Ghazali;Artificial Higher Order Neural Networks for Computer Science and Engineering,2010
4. Research prospective on neural network forecasting.;W. L.Gorr;International Journal of Forecasting,1994
5. Some properties of time series data and their use in econometric model specification