KOSPI200 옵션 거래금액 비율을 이용한 지수 수익률 예측 (Index Return Predictability Using KOSPI200 Option Trading Value Ratio)

서지정보

· 발행기관 : 한국재무학회
· 수록지 정보 : 재무연구 / 36권 / 3호 / 31 ~ 83페이지
· 저자명 : 남태수, 김솔

초록

본 연구는 콜/풋옵션거래금액 비율을 활용하여 KOSPI200 옵션시장과 주가지수시장간 정보 이전 효과를 실증 분석하고, 정보 변수의 가격 발견 기능이 극대화되기 위한시장 조건을 탐색한다. 연구 결과는 다음과 같다. 첫째, 지표 추종 매매 투자 전략에따르면, 콜/풋옵션거래금액 비율이 주가지수에 대해 유효한 가격발견기능을 갖는다.
둘째, 가격 발견 기능의 지속력은 등가격 옵션거래에서 가장 강하게 나타났으며 옵션의 잔존만기가 짧을수록, 주가지수의 변동성이 클수록 가격 발견 효과가 강해지는양상을 보였다. 셋째, 콜/풋 거래금액 비율의 정보력은 기존 문헌에서 활용되던 콜/풋거래량 비율보다 예측 정보력 측면에서 우월한 모습을 보이는 것으로 나타났다. 넷째, 실제 거래비용을 고려하는 경우에는 지표 추종 매매 전략이 초과 수익을 만들지 못하며, 주식시장과 옵션시장 사이의 가격연동이 효율적으로 이루어지고 있음을 확인할수 있다. 마지막으로 KOSPI200 옵션의 거래 승수 인하는 외가격 옵션을 이용한거래 전략의 매매의 성과만 개선시키는 모습을 보이며 거래 승수 인하 정책이 외가격옵션 시장의 효율성을 낮추었다고 볼 수 있다.

영어초록

We empirically analyze the effect of information transfer between the KOSPI200 option market and the stock index market using the call/put option trading value ratio. According to the theoretical framework proposed by Chen et al. (2005), the ratio of call/put trading values is calculated every minute in order to estimate the probability ratio of the index increasing versus decreasing. Then, we implement the indicator-following trading strategy, which buys the index when the benchmark call/put ratio increases more than the pre-defined sensitivity threshold and sells the index when the benchmark call/put ratio decreases more than the sensitivity. Using the calculated daily returns of the indicator-following trading, we test whether the indicator-following trading strategy significantly outperforms the simple buy-hold strategy that takes a long position on the index at the beginning of the market and clears the position when the market closes for each day. If returns of indicator-following trading are significantly larger than the returns of the buy-hold strategy, the empirical results can support the hypothesis that information is disseminated from the KOSPI200 options market to the stock index market, and the call/put ratio can be treated as a predictive information variable. By controlling various environments of indicator-following trading, we investigate market conditions for maximizing the price discovery effect of our information variables: trading timing, trading sensitivity, volatility of the index, time to maturity of the option, and call/put ratios for different moneyness. In addition, we compare the return predictability of our information variables to the call/put option trading volume ratio, take account of transaction costs, and examine the effect of lowering the trading multiplier of the KOSPI200 option.
The results of our study are as follows. First, the indicator-following trading on the total call/put ratio exhibits significantly higher returns than the returns of the buy-hold strategy by 18bp on average per day with the corresponding t-statistic of 7.8, which supports the price discovery feature of our information variable for the KOSPI200 index. Also, the return predictability remains effective for a minute under a 1% significance level. Further, we numerically verify the existence of the local maximum point whose sensitivity maximizes the average return per trading of the indicator-following strategy. As the existence of the local maximum is consistent with the theoretical prediction, the underlying theoretical framework could be justified. Second, the return predictability extends to 2 minutes for the call/put trading value ratio on the at-the-money option, which implies that the persistence of the price discovery stems from the trading information of the at-the-money option. The argument is supported by additional empirical tests under different moneyness ranges. Moreover, the shorter the remaining maturity of the option and the greater the volatility of the stock index is, the stronger the price discovery effect becomes. Third, comparing the returns of indicator-following trading, the return predictability of the call/put trading value ratio is superior to the call/put trading volume ratio used in the existing literature. Fourth, when considering actual transaction costs, the indicator-following trading strategy does not generate excess returns, which indicates efficient and immediate information delivery between the actual option market and the stock index market. Lastly, lowering the trading multiplier of the KOSPI200 option improves only the trading performance of the trading strategy using the out-of-the-money options, and we interpret that the policy hinders the efficiency of the out-of-the-money option market.
Our study contributes to the strands of literature spanning the market microstructure between the KOSPI200 option market and stock index market, the high-frequency trading, and the policy of the Korean option market. By demonstrating the distinguished and superior return predictability of the call/put option trading value ratio compared to the counterpart of trading volume, we argue that the information can flow from the index option market to the stock index market. That is, our results shed light on the unresolved debate on the direction of the information flow between those two markets. Including the significant persistence of the return predictability of the call/put trading value ratio, various empirical findings under different trading conditions support that the KOSPI200 option market leads the stock index market.
Meanwhile, we highlight the importance of using high-frequency data because implementing the high-frequency trading strategy allows us to estimate more precise persistence of return predictability than prior research. Additionally, complementing the extant research reaching conflicting conclusions on the effectiveness of lowering the trading multiplier of the KOSPI200 options, identifying the effect of the policy with the indicator-following trading strategy on call/put trading volume ratio enables us to support the partial efficacy of the policy.

참고자료

· 없음