Sliding Windowed DFNN Pricing Model of Warrants with Complex Terms
Tieke Li · Journal of Zhengzhou University · 2011
A pricing model of warrants with complex terms was proposed based on dynamic fuzzy neural network(DFNN) and Black-Scholes model.In order to get optimized result-parameters of model with fixed-length time series data,a sliding window was set for DFNN.A dynamic adjustment approach of premise parameters was introduced to improve generalization ability of the model.Comparing with RBF pricing model,DFNN showed smaller deviation and higher accuracy in pricing EJiao EJC1.