Date of Award

Spring 1996

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Program/Concentration

Business Administration-Finance

Committee Director

Mohammed Najand

Committee Member

Sylvia C. Hudgins

Committee Member

Christopher Colburn

Abstract

This dissertation is an in depth study of the measurement of pricing biases in futures options, and whether this bias is due to volatility risk premia, market overreaction to public information or information asymmetry. Futures options for thirteen different contracts are used. Additionally, the contracts are from three different marketplaces.

Six hypotheses are tested. The first is whether implied option volatilities from the Black (1976) futures option model is the only significant determinant of the volatility processes of the underlying futures contracts. For this estimation, we use both a GARCH (1,1) model and the Partially Non-parametric model of Engle and Ng (1993). We find that implied option volatility is not the sole significant predictor for of conditional volatility for 10 of the thirteen contracts. For three of the contracts, the implied volatility is insignificant.

Second, we test Stein's (1989) hypothesis of market overreaction. We find that in general, the evidence tends to support the prediction of Stein's hypothesis, though there are important exceptions.

Third, we test Nandi's hypothesis of asymmetric information in the market between traders. We test this by testing the significance of option contract volume on the volatility process. In general, the evidence tends to support Nandi's hypothesis, though again, there are important exceptions.

Fourth, we test the significance of the news response curves as outlined by Engle and Ng (1993). We find that there is little support in the shape and significance of the news response curves to support the presence of volatility risk premiums.

Fifth, we test for differences in the structures of the estimated GARCH models between contracts and model. We find that American based interest-rate futures markets tend to be more highly reactive to innovations than the London (LIFFE) market.

Sixth, we test for the exposure of major commercial banks dealing in futures to a volatility risk premium. While we find evidence that some banks are exposed to our volatility risk premium proxy, the contracts exhibiting significant coefficients generally do not match up with those contracts suspected of harboring volatility risk premiums under the previous tests.

We conclude that there is little empirical support for the presence of a priced volatility risk premium amongst futures and futures options. The presence of pricing biases in such markets seems better explained as being due to information asymmetry or overreaction to news.

DOI

10.25777/10ws-m496

Share

COinS