Aggregate Information Extraction From Option Volume And Its Application

This dissertation attempts to study how to modify the two option-information aggregation methods introduced in Holowczak et al.(2014),namely maturity discount method and strike-price discount method,to read better from the trading volume of Taiwan composite stock index options(TXO).To study how to aggregate information effectively from option volume in an emerging market such as the Taiwan futures market,we should give greater consideration to two more considerations,namely retail trading and investors’ option trading distribution across moneyness,other than market depth,liquidity,leverage and investors5 trading purposes.Retail investors,often considered as noise traders,have traded mainly nearby TXO options with expiration less than one month.Thus,the concentration of retail trading at nearby TXO options may weaken the information role of the said nearby TXO options.Therefore,adopting the modified maturity discounted put-call ratio,which assigns maximum weight to intermediate-horizon options with higher institutions’participation rates,is able to retrieve higher-quality aggregate information from option volume.Furthermore,both institutions and retail investors have traded more at near-the-money TXO options.Therefore,to modify the the strike-price discount method,the weights of in-the-money options and out-of-the-money options should be smaller than those shown in Holowczak et al.(2014)according to the liquidity hypothesis.In addition,we find that there is a dichotomy in the information roles of out-of-the-money options:the information content of their trades is higher(lower)when market volatility increases(falls).Based on this finding,we establish a VIX-adjusted put-call ratio which increases(decreases)the weight of out-of-the-money options when the market VIX is larger(smaller)than its past average level.The empirical results show that it has shown very good predictive ability of large downside market movements.We also examine whether options trading in the TXO market conveys information on Taiwan stock index futures(TX)prices and analyze the informational role of foreign institutions in incorporating information into TX market.When an informed investor places an order in the TXO market,the TXO market makers have to make a counterpart offer.The market makers have cut the delta risk by trading TX.This is how market makers transfer the information in the TXO to the TX market,thus we can conceive of TX order imbalance as being comprised of two parts:the option-induced order imbalance and the option-independent futures order imbalance.We have found that only the option-induced part(OOI)of the total TX order imbalance can predict future TX prices,and there is no price reversal in longer horizon predictive regressions.Furthermore,the OOI calculated from open-buy TXO,defined by Ni et al.(2008),provides incremental predictability.This finding implies that the price predictability of OOI mainly stems from the information asymmetry rather than from liquidity pressure.We conclude further that option transactions from foreign institutions provide the most significant predictability,out-of-the-money option transactions in particular.These empirical results show that option transactions conducted by foreign institutions have played the primary role in conveying the information inherent in the TXO market to the TX market and foreign institutions are more likely to be delta-informed traders.Retail investors,the major players in both the TXO and TX markets,have done almost nothing of significance with regard to TXO information transmission into the TX market,with the exception of some near-the-money and out-of-the-money options.