overed call writing is a popular investment strategy among both individual and C institutional investors. The introduction of index option contracts created a new vehicle for those programs. Effective maintenance of a covered index position is potentially complicated by the requirement that the underlying stock portfolio be adjusted for both dividend reinvestment and changes in the composition of the index. The costs associated with these requirements could be material, thereby reducing the profitability of the strategy. One potential means of mitigating the impact of those costs would be to select a representative subset of securities in the index as a proxy for the entire portfolio.
In this paper, we analyze the investment performance of alternate procedures for the selection of a cross-hedge position. Both stepwise regression and value-ranking techniques are shown to be effective in the identification of an optimal proxy portfolio.
I. INTRODUCTION
Following the inception of option trading on the Chicago Board Option Exchange (CBOE) in April 1973, the practice of writing covered call options has become popular among individuals, pension funds, and corporations. The attraction of the strategy appears to lie with the potential for the call premium to cushion downside movements in the long stock position. A number of studies , . Pounds (1978)l havc examined the risk-return characteristics of writing individual covered calls. Generally speaking, those studies have provided mixed results on the impact to realized returns, while demonstrating a definite tendency for the call writing program to reduce return variability.
Call writing programs are employed for a wide variety of reasons. Corporate cash managers often employ covered call positions as a means of capturing dividends on a hedged basis [Brown and Lummer (1984,1986) and Zivney and Alderson (1986)l. Writing call options against a long position in the underlying stock is also appealing to investors with a strong aversion to regret [Shefrin and Statman (1985a, 1985b)l; or to those who use it as a mechanism for converting principal to "income."
We are grateful to Ketan Pate1 for his very capable research assistance, and to the participants in the Finance workshop at Texas A&M University for helpful comments and suggestions. All remaining errors are our own.