A New Method To Quantify Corss-correlation Between Two Nonstational Time Series

Econophysics is an interdiscipline of applied mathematics, statistical physics and finance. Analyzing the correlation of financial time series is a hot field of research. To detect autocorrelation of single time series, C. Peng proposed Fluctuation Analysis (FA) algorithm and Detrended Fluctuation Analysis (DFA) method during the study of DNA sequences.H. Stanley extended DFA to study the cross-corrlation between two nonstationary time series and proposed a Detrended Cross-Correlation Analysis (DCCA) algorithm.In this thesis, we employ the core idea of fluctuation analysis algorithm; develop a new algorithm to detect the cross-correlation between two nonstationary time series:Fluctuation Cross-Correlation Analysis (FCCA) algorithm. Two different numerical algorithms are used to generate different Hurst exponent sequences to compare the new algorithm to the current existing ones. Finally, we apply the methods to study the influences between different stock markets.