Nonlinear Dynamics in CEE Stock Market Indices

We investigate the existence of nonlinear patterns in the dynamics of the main stock index returns in selected CEE countries. We use several types of nonlinear tests, among which White test, runs test, Teraesvirta's neural network test and BDS test as well as a series of techniques borrowed from chaos theory, like maximum Lyapunov exponent or correlation dimension. We discuss the implications of the results with respect to the efficient market hypothesis for the selected financial markets.