|
Activity Number:
|
472
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 1, 2007 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Business and Economics Statistics Section
|
| Abstract - #309444 |
|
Title:
|
The Economic Value of Intelligently Subsampled Realized Covariance Estimation of Asynchronous and Noisy High-Frequency Data
|
|
Author(s):
|
Lada Kyj*+ and Katherine Ensor and Barbara Ostdiek
|
|
Companies:
|
Rice University and Rice University and Rice University
|
|
Address:
|
PO Box 1892 MS 138, Houston, TX, 77251,
|
|
Keywords:
|
Realized Covariance ; Market Microstructure ; Asynchronous Observations ; Subsampling ; Volatility timing ; High-frequency Data
|
|
Abstract:
|
Assessing the economic value of portfolios constructed with more precise estimates of covariance is of great interest in finance. We present a realized covariance estimator that incorporates cross-market tick-matching and intelligent subsampling. These features of the estimator offer the potential for improved performance in the presence of asynchroneity and market microstructure noise. Specifically, tick-matching preserves information when arrival structures are nonsynchronous, intelligent sampling reduces microstructure-induced noise, and averaging reduces variance. We compare the performance of this estimator with prevailing methodologies in a simulation study and by assessing out-of-sample volatility-timing portfolio optimization strategies. Results show that our estimator has smaller MSE, smaller bias, and greater economic utility than prevailing methodologies.
|
