Use this URL to cite or link to this record in EThOS:
Title: Essays in empirical finance
Author: Xu, Qi
ISNI:       0000 0004 5916 1183
Awarding Body: University of Warwick
Current Institution: University of Warwick
Date of Award: 2016
Availability of Full Text:
Access from EThOS:
Access from Institution:
This thesis consists of three papers in the area of empirical finance. Chapter 2 investigates the role of realized jumps detected from high frequency data in predicting future volatility from both statistical and economic perspectives. We show that separating jumps from diffusion improves volatility forecasting both in-sample and out-of-sample. Moreover, we show that these statistical improvements can be translated into economic value. We find a risk-averse investor can significantly improve her portfolio performance by incorporating realized jumps into a volatility timing based portfolio strategy. Chapter 3 investigates the use of high frequency data in large dimensional portfolio allocation. We consider the use of high frequency data beyond the estimation of the realized covariance matrix. Portfolio strategies using high frequency data in measuring and forecasting univariate realized volatility can generate statistically significant and economically tangible benefits compared to low frequency strategies. Moreover, using high frequency data to separate realized volatility into different components and construct realized higher moments can also enhance portfolio performance. Strategies using upside and downside volatility components and using realized skewness can deliver incremental economic benefits over the strategy using total realized volatility alone. Chapter 4 investigates the pricing of volatility risks in currency markets. Firstly, we show that pricing volatility risk can be understood by pricing some of its components. We find that diffusive volatility dominates jump volatility in pricing carry trade returns, while jump volatility is important to explain the joint cross-section of carry trade and momentum returns. Both short run and long run components are priced, and the short run component is more important in general. Secondly, we suggest that factors similar to volatility in identifying bad states, i.e. volatility of volatility and cross sectional volatility are also priced in currency returns and they cannot be fully subsumed by conventional volatility risks.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: HG Finance