Journal of Fundamentals of Renewable Energy and Applications
Open Access

Fractional co-integrated vector auto regression (FCVAR) methods: An application between energy and stock markets

2^nd International Congress and Expo on Biofuels & Bioenergy

August 29-31, 2016 Sao Paulo, Brazil

Nadhem Selmi

University of Sfax, Tunisia

Posters & Accepted Abstracts: J Fundam Renewable Energy App

Abstract :

Market series� univariate dependencies, both objective and chock persistent can be effectively, explained by long-memory fractionally integrated models. For now, the ex-post difference, or the variance swap payoff reflecting the return bearing reward, displays far less persistent dynamics. In this respect, a process based on inference a fractionally co-integrated (or co-fractional) vector autoregressive process will be employed in this work while relying on an initial value conditional Gaussian likelihood. Indeed, applying some developed countries data oil-price and stock-market coupled with frequency domain methods, we undertake to dislocate the series into various parameters. Actually, our reached conclusions are discovered to be consistent with generalized long-run processes and help well provide some plausible explanations of the major why classical efforts of reasons lying behind the classical efforts� failure to establish a naive return-series relationship. In addition, a fractionally co-integrated vector auto-regression (FC-VAR) has also been estimated as part of this work. In effect, the long-run equilibrium relationship process established between the five indices has yielded a non-trivial return predictability over monthly horizons, which sustains the fact that a co-integrating relationship among the five indices turns out to help greatly in measuring proxies pertaining to for the financial market-satisfies uncertainty.

Biography :

Email: nadhem.selmi@yahoo.fr