Nonlinear Time Series Analysis

Recent advances in signal processing and nonlinear time series analysis raise the possibility of improving the power of financial forecasting techniques. For example, methods such as wavelet decomposition, support-vector machines, nonparametric regression, and artificial neural networks have all demonstrated superior performance in many engineering applications. In this project, we propose to apply and extend such methods to the financial time series in a series of studies, focusing on various markets and applications.

Relevant Publications and Preprints:

• Lo, A., Mamaysky, H., and J. Wang, 2000 “Foundations of Technical Analysis: Computational Algorithms, Statistical Inference, and Empirical Implementation,”  Journal of Finance 55,1705-1765.