Abstract
This paper uses the cross bicorrelation methodology, which can capture nonlinear trascendence periods through window functions and third-order moments. It applies to the return of four sets of commodities of coffee traded on the New York market (Arabica Colombian, mild Arabica, Arabica Brazilian and Other Arabicas), during the 20/06/1997 - 27/10/2010 period. The results conclude that there is a cross bicorrelation among the four series, with Brazilian type coffee being the leader and a lower bicorrelation with other Arabicas. This complicates decisions for investors in such series.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have