幾乎隨機優越方法下的選擇權價格隱含邊際效用比值上下界

Abstract

本文提出一個可利用選擇權價格估計一階幾乎隨機優越偏好參數的方法，此偏好參數可被一對一轉換為邊際效用比值之上下界、或絕對風險趨避函數和之上下界。利用1996年1月至2023年8月之S&P 500指數選擇權價格資料，本文得到之偏好參數估計值與文獻中從其他資料來源得到之估計值相似。本文亦測試選擇權天期(maturity)與價性(moneyness)對此偏好參數估計值的影響，在實證結果中，對於投資在接近價平或較長天期選擇權之投資人群體，實證結果傾向支持其風險偏好較不極端。

A method implying the maximum allowed almost first-degree stochastic dominance (AFSD) preference parameter with option prices is proposed, where the AFSD preference parameter can be one-to-one transformed as the marginal utility ratio bounds or the bounds of the sum of the absolute risk aversion function. The preference parameter is estimated with S&P 500 index option data from January 1996 to August 2023, and the estimates are similar to those from other data sources. Maturity and moneyness effects on the preference parameter are further examined, and evidence supports that investors investing in near at-the-money or long maturity options tend to have less extreme risk preferences.