An Empirical Study Of The Fu Dan RMB Exchang Rate Indices Based On Fractalanalysis

With the rapid development of science and technology,the nonlinear science has become the forefront research of the scientists to study the complexity problem.In the financial markets at present,various data change extremely complex,such as stock prices,exchange rate indices,the prices of gold daily changes and so on.The domestic and foreign scholars began to research the regularity of data,in order to find the way to forecast the financial market.Some scholars found that,the change of financial data is not disorderly and its change rule is not random,but it has certain relevance.So we should look for a method that can describe and predict financial market change law,besides that,the method which can make more accurate estimation and prediction is of vital importance.Currency market is an important and necessary part of the financial market.Based on the nonlinear theory in this paper,the intelligent calculation method and fractal theory were applied to study the currency market and measure the risk size of currency market.The characters of fluctuation of exchange rate indices were analyzed and the prediction method are analyzed and compared.The structure of this paper is as follows:The first chapter introduces the background and significance,research status at home and abroad and the research content of this article,the innovations of this article are pointed out.The second chapter introduces the theory of fractal and the method of fractal analysis.R/S analysis method,DFA method and MF-DFA method are important roles.In the third chapter,the methods of the multifractal spectrum and the multifractal detrended fluctuation analysis are applied to analyze volatility characteristics of the time series of Fudan RMB exchange rate indices,and also applies VaR model to measure their risk.The empirical results show that the returns of the Fudan RMB exchange rate indices possess multifractal characteristics and the multifractal strength of real effective exchange rate great slightly than that of nominal exchange rate.The VaR model is suitable to measure their risk,and it is found that the VaR values are different under the different confidence levels.The higher the confidence level,the greater the risk of the measuring market.The fourth chapter makes a research to the prediction of Fudan RMB real effective exchange rate indices.NAR neural network model and fractal interpolation model are established,and the prediction models are analyzed and compared.Results show that the fractal interpolation model is better than the NAR neural network model of prediction effect.The fifth chapter is summary and outlook.