Testing for Nonlinearity and Chaoticity in Exchange Rate Time Series

摘要

The nonlinearity and chaoticity of exchange rate time series is investigated with complex systems theory. The data set that has been used in this analysis consists of 4 daily exchange rate recorded. The BDs test and surrogate data method indicate that the exchange rate time series of Canadian Dollars to United States Dollar (CD/USD), Japanese Yen to United States Dollar (JY/USD) and United States Dollar to British Pound (USD/BP) exhibit nonlinearity, while the exchange rate time series of United States Dollar to EURO (USD/EURO) is linear. The positive largest Lyapunov exponents have provided evidence for the possibility of deterministic chaos in the daily exchange rate time series of CD/USD, JY/USD and USD/BP. The research has important theoretical and practical significances for phase-space reconstruction, modeling and prediction of the exchange rate time series.