Essays on Option Implied Information

Abstract

This thesis aims to understanding the option implied information, including political event risk, after-hours option trading, and Ross recovered probability. It consists of three essays. The first essay shows that the option market can ex ante detect and quantify the effects of political event risk. Focussing on the 2016 UK referendum on EU membership, we find that the Risk-Neutral Distribution extracted from GBPUSD futures options whose expiry spans the referendum date becomes bimodal and the Implied Volatility curve exhibits an unusual W-shape. To the contrary, the corresponding effects for FTSE100 are found to be very limited. The large swings in expectations regarding the event outcome during the referendum night allow us to observe the counterfactual and validate the ex ante information revealed in the option market. The second essay documents the trading characteristics of the option market in extended trading hours and explores the intraday intertemporal decisions of liquidity and informed traders. The option market exhibits low liquidity, low trading activities, high transaction costs, and a high probability of informed trading in extended trading hours. However, the introduction of extended trading hours enhances market quality by decreasing the quoted and effective bid-ask spreads in the following regular trading hours with decreased adverse selection costs. Moreover, option prices in extended trading hours are informative for the following regular trading hours in terms of index level and realized volatility. The third essay presents the importance of an interest rate condition in Ross recovery. The term structure of interest rates explains the differences between approaches that follow the original Ross recovery theorem. A flat term structure of interest rates results in a Ross recovered probability distribution identical to the risk-neutral probability distribution in Ross recovery. After considering interest rates with a market example, empirical evidence still shows Ross recovered probabilities are close to the risk-neutral probabilities. We propose new estimation approaches to reflect the requirements in Ross recovery without imposing additional assumptions. Ross recovery with a short transition period implies a nonnegative matrix root for the transition matrix with a long transition period. Different least squares estimations are not equivalent when there is no unique and exact fitting with the market spot state prices. A sparse spot state price surface probably results in a relatively stable pricing kernel in Ross recovery. By taking into account the implicit requirements in Ross recovery, we propose a recovery estimation labeled Ross Power with a multi-period interest rate condition. This approach provides insights into the relationship among Ross recovered, generalized recovered, and risk-neutral probabilities.

Date of Award	1 Aug 2022
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Alexandros Kostakis (Supervisor) & Yoichi Otsubo (Supervisor)