Abstract
We present a methodology to estimate the covariance of two time series when they are sampled from continuous semimartingales at general stopping times in a nonsynchronous manner. Approximation error bounds being explored, the estimators are shown to be consistent as the size of the sampling intervals tends to zero. The methodology is easy to be implemented with potentially broad applications, especially in financial modeling and analysis involving high-frequency transaction data. The results generalize those recently obtained by obtained by Hayashi and Yoshida (2005, Bernoulli 11(2):359-379)
| Original language | English |
|---|---|
| Pages (from-to) | 93-106 |
| Number of pages | 14 |
| Journal | Statistical Inference for Stochastic Processes |
| Volume | 11 |
| Issue number | 1 |
| DOIs | |
| Publication status | Published - 2008 Feb 1 |
Keywords
- Consistency
- Discrete-time sampling
- High-frequency data
- Nonsynchronous trading
- Quadratic variation
- Realized covariance
- Semimartingale
- Stopping time
ASJC Scopus subject areas
- Statistics and Probability
Fingerprint Dive into the research topics of 'Consistent estimation of covariation under nonsynchronicity'. Together they form a unique fingerprint.
Cite this
Hayashi, T., & Kusuoka, S. (2008). Consistent estimation of covariation under nonsynchronicity. Statistical Inference for Stochastic Processes, 11(1), 93-106. https://doi.org/10.1007/s11203-007-9009-9