콜옵션과 풋옵션 간의 정보수렴과정

This study empirically tests some hypothesis about the information convergence processes between the call and put options market under limits to arbitrage, by using the methodology designed in terms of options market participants as well as econometric models, and then, analyzes whether considering the cross effects of implied informations into the information convergence processes can improve out-of-sample pricing performance. The major conclusions drawn from the empirical results are summarized as follows. First, the implied information reflected only in call (put) prices gradually have a decreased effect on a put (call) prices during three or four trading days. Second, the information convergence function of the call options market may be more efficient than the put options market. Specifically, the implied information reflected only in call (put) prices converge more quickly on the put options market before independent information of put options is fully reflected in call options prices. And, the degree of a complexity, irregularity and unpredictability in time series of the information implied in the call options prices, estimated by using the approximate entropy method proposed by Pincus (1991), is shown to be higher than those in put options prices. Third, the robustness testing which can control the effect of the decision errors be caused by variables except implied information in forecasted option prices reach the same conclusions about the information convergence processes between call and put options market. And the estimates of vector autoregression used to capture the linear interdependencies among volatilities implied in call and put options also support the hypothesis. Last, it is indeed useful to jointly model call and put implied volatilities, in which the cross effects of implied informations into the information convergence processes can be captured, to improve out-of-sample pricing performance up to two trading days, regardless of different weighting schemes of call and put implied volatilities. Especially, the simple scheme in which weighting are monotonically decreasing (increasing) in the forecasting days is not only more useful but also more easy to use than ad-hoc Black and Scholes model which is one of the most widely used option valuation procedures among practitioners.