Solve fractal dimension of Shanghai stock market by RBF neural networks

The actual financial time series is random walk process which is biased, and has significant fractal features and long-term memory effect. Research results also show the existence of low dimension chaos in stock market. Fractal dimension is the salient features of stock market chaos, and the next integer larger than correlation dimension is the number of the independent system variable which is required by setting up the system dynamics model. In this paper, we get the fractal dimension of Shanghai stock market through function approximation algorithm of RBF neural networks and discuss the influence of sample size and growth trend. It provides a measurement method for fractal dimension.