The hedging effectiveness of futures contracts : comparison of the Mean Gini and Mean Variance frameworks

Van Niekerk, Leon (2003-12)

Thesis (MBA)--Stellenbosch University, 2003.

Thesis

ENGLISH ABSTRACT: When hedging with futures contracts the hedge ratio is one of the fundamental figures needed to set up a successful hedging strategy. The Mean Variance framework has been used for some time to calculate hedge ratios for this exact purpose. Certain assumptions are implicit in the Mean Variance framework such as the assumed normality of returns and assumed quadratic utility functions of investors. The validity of these assumptions has been questioned in the literature. The Mean Gini framework is first and second order stochastically dominant implying that there is no assumption regarding the return process or the utility function of investors. This study compares these two frameworks regarding the hedge ratios generated by each and their subsequent hedging effectiveness. The results indicate that neither one of the two frameworks generate the most effective hedge ratios all the time. The Mean Gini framework is, however, preferred above the Mean Variance framework for a significant number of futures contracts evaluated. It can therefore be concluded that making use of the Mean Variance framework for all futures contracts would have resulted in numerous ineffective hedging situations.

AFRIKAANSE OPSOMMING: Wanneer verskansing met termynkontrakte gedoen word is, die verskansingsverhouding een van die fundamentele veranderlikes waaroor 'n besluit geneem moet word. Vir 'n geruime tyd word die Minimum Variansie verskansingsverhouding vir hierdie doel gebruik. Implisiet in die Minimum Variansie raamwerk is die aannames dat opbrengste normaal verdeel is en dat die beleggersnutsfunksie kwadraties van aard is. Die geldigheid van hierdie aannames het reeds heelwat kritiek in die literatuur ontlok. Die Gemiddelde Gini raamwerk is eerste en tweede graads stochasties dominant, wat impliseer dat geen aannames aangaande die opbrengs of die beleggersnutsfunksie gemaak word nie. Die studie vergelyk beide raamwerke rakende die verskansingsverhoudings deur elk gegenereer en die gepaardgaande verskansingdoeltreffendheid. Die resultate toon dat nie een van die twee raamwerke vir alle kontrakte die mees effektiewe verskansingsverhouding genereer nie. Die Gemiddelde Gini raamwerk word egter vir 'n beduidende aantal van die termynkontrakte bestudeer bo die Minimum Variansie raamwerk verkies.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/53640
This item appears in the following collections: