Non-linear modelling of bivariate comovements in asset prices

The phenomenon of comovements among asset prices has received a lot of attentions for several reasons. The increasing interest in this topic has been the reason of the production of a large number of contributions. In this paper we propose an investigating methodology for the non-linear modelling of bivariate comovements. Our methodology leaves the ones presented in the recent literature. In fact, our approach, which is articulated in three steps, allows the evaluation and the statistical testing of non-linearly driven comovements between two given random variables. Moreover, when such an (unknown) bivariate dependence relationship is detected, our approach allows also to provide a polynomial approximation of it. Finally, we apply our three-steps methodology to some energy asset prices time series traded in the U.S.A.. The goodness of the results is encouraging given the novelty of the proposed investigating approach.

In this paper we propose an investigating methodology for the non-linear modelling of bivariate comovements. Our approach leaves the ones presented in the recent literature. In fact, our approach allows the evaluation and the statistical testing of non-linearly driven comovements between two given random variables. Moreover, when such an (unknown) bivariate dependence relationship is detected, our approach allows also to provide a polynomial approximation of it. Finally, we apply our methodology to some energy asset prices time series traded in the U.S.A.. The goodness of the results is encouraging given the novelty of the proposed approach.