Non-expected utility hedge ratios

The purpose of my dissertation is to compute and investigate the properties of optimal hedge ratios for futures and options on futures using non-expected utility models. This is a valuable mission because expected utility fails to accurately predict human behavior in a wide range of settings including the optimal trading of derivatives. The application of non-expected utility to hedging models is, for the most part, new to the hedging literature. The non-expected utility models that will be considered are weighted expected utility, rank dependent utility, and cumulative prospect theory.;Although the use of expected utility is widespread and unquestioned by many economists there are significant problems with it. First and foremost, expected utility does not appear to describe how agents actually behave. There is ample experimental evidence that expected utility theory breaks down when agents are faced with very large payoffs, positive or negative. While it is unrealistic to expect any theory to completely explain human behavior the problems with expected utility are large enough to justify the exploration of alternatives. Matthew Rabin and Richard Thaler (2001) make this argument persuasively in a recent edition of the "Anomalies" series in the Journal of Economic Perspectives. In their words: "Conventional expected utility theory is simply not a plausible explanation for many instances of risk aversion that economists study.";In particular the failings of expected utility are relevant to the study of hedging. Understanding agent behavior in the face of catastrophic loss (or gain) is necessary to construct optimal hedging strategies or to understand why and how much agents hedge. Most theoretical hedging models are based on expected utility, so if expected utility is not an adequate model of human behavior then existing hedging models must be re-examined within the context of non-expected utility. Computing non-expected utility hedge ratios may also shed light on why observed hedging behavior is not consistent with expected utility theory; agents make much less use of available futures markets than conventional theory suggests they should. Expected utility models also tend to predict little or no role for options despite the fact that there are many vibrant options markets.;Problem statement. The purpose of my dissertation is to compute hedge ratios for futures and options on futures using non-expected utility models. Expected utility fails to explain human behavior in the face of large gains and large losses making hedging models based on expected utility suspect. Calculating hedge ratios based on the alternatives to expected utility is an obvious step in understanding optimal and observed hedging behavior. The research represents an original contribution to the risk management literature.