Drift and volatility

Liao, Kefu 2024. Drift and volatility. PhD Thesis, Cardiff University. Item availability restricted.

Preview	PDF - Accepted Post-Print Version Download (5MB) | Preview
	PDF (Cardiff University Electronic Publication Form) - Supplemental Material Restricted to Repository staff only Download (289kB)

Abstract

High-frequency stock price dynamics are conventionally modelled by three components: volatility, drift, and jumps. Volatility depicts how greatly an asset’s prices swing around the mean price. Drift describes the movement of the mean price. Jumps refer to rare, significant, and sudden price changes that are too large to be explained by volatility alone. This thesis investigates the econometrics of volatility, jumps, and drift in high-frequency stock prices and the implications of these components for the stock markets. Chapter 1 contains the introduction of this thesis. Chapter 2 reveals non-negligible drift-related finite sample biases in the estimation of good volatility, bad volatility, and signed jumps. This chapter suggests a modified estimation method that significantly reduces these biases. The empirical evidence indicates that the asymmetric impacts of good and bad volatility and the asymmetric effects of signed jumps in volatility forecasting, as found in the literature, are almost exclusively due to the influence of the measurement bias of these variables on future volatility. Chapter 3 is subdivided into two parts. The first part focuses on the measurement of the occurrence, size, and intensity of drift bursts. The empirical results reported in the second part imply that drift bursts do not impact realized variance but explain the implied variance and variance risk premium. Chapter 4 first demonstrates that applying a coexceedance criterion to a univariate drift test proposed by recent studies is feasible for detecting stock codrift variations. I show stock codrift variations are significantly associated with market drift bursts. I also find stock codrift variations have a significant and positive impact on market volatility. Models exploiting this effect lead to significantly better in-sample and out-of-sample market volatility forecasts. Chapter 5 is the conclusion of this thesis.

Item Type:	Thesis (PhD)
Date Type:	Completion
Status:	Unpublished
Schools:	Business (Including Economics)
Date of First Compliant Deposit:	20 January 2025
Date of Acceptance:	20 January 2025
Last Modified:	20 Jan 2025 15:56
URI:	https://orca.cardiff.ac.uk/id/eprint/175404

Actions (repository staff only)

Edit Item