Duration-Based Volatility Estimation | Hanlon Financial Systems Center

Duration-Based Volatility Estimation

Duration-Based Volatility Estimation

seminar date: 
Thursday, November 17, 2016 - 6:15pm
seminar location: 
BC122
Dobrislav Dobrev, Senior Economist at the Federal Reserve Board
Abstract: 

We develop a novel duration-based approach to estimating the integrated variance (IV) of a general jump-diffusion with stochastic volatility robustly to both jumps and market microstructure noise. Our approach builds on the relationship between the speed (distance traveled per fixed time unit) and passage time (time taken to travel a fixed distance) of the Brownian motion. In particular, we exploit that measuring the waiting times to observe economically significant threshold crossings automatically adapts to the inherent variations in the local arrival rates of information as an important distinction from return-based estimation methods relying on a fixed sampling frequency throughout the day. Data-based ranking against commonly used benchmark IV estimators indicates consistently higher accuracy rank of our duration-based estimators across the majority of Dow Jones 30 stocks. 

 

Bio: 

Dobrislav Dobrev is a Senior Economist at the Federal Reserve Board where most recently he has been recognized for his contributions to the Joint Staff Report “The U.S. Treasury Market on October 15, 2014” and related Liberty Street Economics Blog posts. Dr. Dobrev’s primary research focus is on financial econometrics in data-rich environments. He has contributed to the development of a variety of methods for measuring volatility, jumps, and co-movements in high frequency financial market activity as well as new techniques for latent factor extraction with emphasis on robust inference in finite samples and applications to risk measurement and forecasting. He is currently working on problems related to the cross-market impact of high frequency trading, proper identification and attribution of major market moving events, low-frequency data modeling incorporating high-frequency data aggregates, as well as forecasting of macro-financial variables in relation to stress-testing. His quantitative risk management experience goes back to 1999 when he developed an innovative, at the time, stress-testing framework for managing the foreign reserves and net worth of the Bulgarian National Bank operating under a currency board. He holds a PhD in Finance from Northwestern University’s Kellogg School of Management and an MSc in Applied Mathematics from Sofia University, Bulgaria.