The effect of option trading volume on underlying stock price predictability
Kilpeläinen, Lassi (2018)
Kilpeläinen, Lassi
2018
Kuvaus
Opinnäytetyö kokotekstinä PDF-muodossa.
Tiivistelmä
The purpose of this thesis is to study effects of option trading volume on stock price predictability. More specifically, the paper examines whether option trading volume contains information that forecasts underlying stock returns. The study challenges the efficient market hypothesis and examine whether information flows firstly into options until it is fully priced in the underlying share instead of simultaneous adaptation.
The sample consists of companies that are traded on the derivative markets of the London LIFFE and the EUREX Germany, hence the study concentrates mainly on European firms. The data covers period from January 2000 to September 2017. Additionally, different sample splits around the financial crisis (FC) are considered in order to test whether unusual trading volume during the FC affect returns. Return predictability is tested by sorting companies into quintiles by their put-call (put-to-call option trading volume, P/C) or option-to-share (option-to-stock trading volume, O/S) ratios and measuring their future returns at daily and weekly levels.
The findings of the study indicate that low P/C and O/S ratios predict positive next day and next week returns while predictability of weekly returns of the O/S are the strongest. High ratios do not contain any significant predictive power. However, the difference between extreme O/S quintiles (Q5–Q1) at weekly level has found to be highly significantly negative which implies that high O/S level is more bearish signal. Results are robust over time and regressed on four-factor model which cannot fully explain the abnormal returns. The results propose that options are chosen over stocks to trade bad news which may be due to option’s higher leverage opportunities and cost-effectiveness of synthetic short positions.
This study contributes on the existing literature not only by supporting the prior findings of option volume predictability but also by showing that the predictive power is stronger in small cap and value stocks.
The sample consists of companies that are traded on the derivative markets of the London LIFFE and the EUREX Germany, hence the study concentrates mainly on European firms. The data covers period from January 2000 to September 2017. Additionally, different sample splits around the financial crisis (FC) are considered in order to test whether unusual trading volume during the FC affect returns. Return predictability is tested by sorting companies into quintiles by their put-call (put-to-call option trading volume, P/C) or option-to-share (option-to-stock trading volume, O/S) ratios and measuring their future returns at daily and weekly levels.
The findings of the study indicate that low P/C and O/S ratios predict positive next day and next week returns while predictability of weekly returns of the O/S are the strongest. High ratios do not contain any significant predictive power. However, the difference between extreme O/S quintiles (Q5–Q1) at weekly level has found to be highly significantly negative which implies that high O/S level is more bearish signal. Results are robust over time and regressed on four-factor model which cannot fully explain the abnormal returns. The results propose that options are chosen over stocks to trade bad news which may be due to option’s higher leverage opportunities and cost-effectiveness of synthetic short positions.
This study contributes on the existing literature not only by supporting the prior findings of option volume predictability but also by showing that the predictive power is stronger in small cap and value stocks.