Several methods have been developed for filtering seasonal structures and extreme returns in financial and economic series. The theoretical support for this approach is rather questionable since it focuses on the effects of shocks on prices and not on their sources. The non-proportional...

In this paper we introduce a new model, based on the synthesis of conditional copulas and Gaussian graphical models under a copula -- vine framework. The use of the copula vine permits each pair between the market and a stock to have their own dynamics. In that case the asset keeps its...

In practice it is very difficult to distinguish between stochastic and chaotic non-linearity. It is also difficult to identify non-linear structures in time series when the hidden dynamics are complex. In this paper we provide evidence that in presence of specific high-dimensional underlying...

Financial returns series typically exhibit excess kurtosis and volatility clustering. The GARCH often is applied to describe these two stylized facts. Nevertheless, in applications of this model to stock returns series it is usually found that it cannot capture all excess kurtosis and high...

The purpose of this paper is to propose a version of causality testing that focuses on the role of the sign of the returns on the causality results. We replace the traditional VAR specification used in the Granger causality test by a discrete-time bivariate noisy Mackey-Glass model. Our test...

The purpose of this paper is to propose a version of causality testing that focuses on the role of the sign of the returns on the causality results. We replace the traditional VAR specification used in the Granger causality test by a discrete-time bivariate noisy Mackey-Glass model. Our test...

Several recently developed chaotic forecasting methods give better results than the random walk forecasts. However they do not take into account specific regularities of stock returns reported in empirical finance literature, such as the calendar effects. In this paper, we present a method for...