On Friday 11 January 2013 Karim Bannouh will defend his PhD thesis entitled 'Measuring and Forecasting Financial Market Volatility using High-Frequency Data'. Supervisor is Professor Dick van Dijk and co-supervisor is Martin Martens. Other members of the Doctoral Committee are Professor Richard Paap (Erasmus University Rotterdam), Professor Michael McAleer (Erasmus University Rotterdam), and Professor Peter Boswijk (University of Amsterdam).
Time and location
The PhD defence will take place in the Senate Hall of Erasmus University Rotterdam and will start at 09.30 hrs.
About the dissertation
The importance of accurate measures and forecasts of financial market risk is exemplified by recent crises such as the bursting of the IT Bubble, the US subprime mortgage crisis and Europe’s sovereign debt debacles. Given the rapidly increasing availability of intraday asset price data and increasing computing power several researchers have illustrated how the use of intraday data can improve risk management. By making efficient use of high-frequency data the precision of risk estimates and forecasts improves substantially compared to estimates and forecasts that are traditionally based on low-frequency data. Risk measures and models based on high-frequency data are more adaptive to recent market dynamics and less impacted by structural breaks than traditional approaches.
In his PhD dissertation entitled Measuring and Forecasting Financial Market Volatility using High-Frequency Data, Karim Bannouh develops financial market risk estimators that improve or compete with existing popular volatility estimators that are based on high-frequency asset price data.
His first study proposes a refined heuristic bias-adjustment for the realised range-based volatility estimators with the objective of further decreasing the impacts of microstructure frictions. Karim Bannouh’s second study introduces the realised co-range, an efficient and innovative estimator of the covariance between asset returns based on intraday high-low price ranges. The mixed-frequency factor model (MFFM), an original and highly efficient vast-dimensional covariance matrix estimator, is introduced in the third study. The MFFM is a novel approach for estimating and forecasting multivariate volatility by combining the use of high- and low-frequency data with a linear factor structure. As such, Bannouh’s study is the first study in the volatility literature that proposes a covariance matrix estimator which can be implemented for portfolios containing a large number of assets whilst making efficient use of high-frequency data. Vast-dimensional asset portfolios are typically held by pension funds, insurance companies, banks and fund managers. The global market indexes that are used to evaluate the performance of fund managers are often based on hundreds or even thousands of stocks, and for these practical reasons is it is key that a covariance estimator is able to handle such large dimensional asset portfolios.
About Karim Bannouh
Karim studied financial econometrics at Erasmus University Rotterdam. In 2007 he started as a PhD candidate at the Econometric Institute (EI) and Erasmus Research Institute of Management (ERIM) at Erasmus University Rotterdam. During his PhD track his work was presented at several international conferences in Aarhus, Bergamo, Geneva, London, Oxford and Stanford. The article version of Chapter 3 is published in the Journal of Financial Econometrics. After his PhD track, he worked as a quantitative analyst at the firm-wide ING Group Model Validation department focusing on financial market risk and trading risk models across asset classes. In May 2011 he joined Saemor Capital, a market neutral hedge fund that uses quantitative strategies, as a portfolio manager with risk management, asset allocation and market timing as his specific fields of interest.
Abstract of Measuring and Forecasting Financial Market Volatility using High-Frequency Data
This dissertation consists of three studies on the use of intraday asset price data for accurate measurement and forecasting of financial market volatility. Chapter 2 proposes a refined heuristic bias-correction for the two time scales realised range-based volatility estimator in the presence of bid-ask bounce and non-trading. The merits are illustrated through simulations and an empirical forecasting application. Chapter 3 introduces a novel approach for estimating the covariance between asset returns using intraday high-low price ranges. The realised co-range estimator compares favourably to the realised covariance for plausible levels of microstructure noise and non-synchronous trading. The estimator is successfully implemented in a volatility timing strategy that deals with constructing mean-variance efficient asset allocation portfolios from stock, bond and gold futures. Chapter 4 introduces a mixed-frequency factor model for vast-dimensional covariance estimation. This original approach combines the use of high- and low-frequency data with a linear factor structure. We propose the use of highly liquid ETFs -- that are essentially free of microstructure frictions -- as factors such that factor covariances can be estimated with high precision from ultra-high-frequency data. The factor loadings are estimated from low-frequency data to bypass the potentially severe impacts of noise for individual stocks and to circumvent non-synchronicity issues between returns on stocks and liquid factors. Theoretical, simulation, and empirical results illustrate that the mixed-frequency factor model is excellent, both compared to low-frequency factor models and to popular realised covariance estimators based on high-frequency data.