On covariance estimation for high-frequency financial data

Takaki Hayashi, Nakahiro Yoshida

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

We consider the problem of estimating the covariance/correlation of two diffusion prices that are observed at discrete times in a nonsynchronous manner. The modem, popular approach in the literature, "realized" estimator, which is based on the sum of cross-products of intra-day log-price changes measured on regularly-spaced intervals over a day, is problematic because choice of regular interval size and data interpolation scheme may lead to unreliable estimation. We present a new estimation procedure recently proposed by [1], which is free of such "synchronization" of data, hence, free from biases or other problems caused by it. In particular, our estimators are shown to have consistency as the observation frequency (or the market liquidity) tends to infinity, which is not possessed by the realized estimators.

Original languageEnglish
Title of host publicationProceedings of the Second IASTED International Conference on Financial Engineering and Applications
EditorsM.H. Hamza
Pages282-286
Number of pages5
Publication statusPublished - 2004
Externally publishedYes
EventProceedings of the Second IASTED International Conference on Financial Engineering and Applications - Cambridge, MA, United States
Duration: 2004 Nov 82004 Nov 10

Other

OtherProceedings of the Second IASTED International Conference on Financial Engineering and Applications
CountryUnited States
CityCambridge, MA
Period04/11/804/11/10

Fingerprint

Modems
Interpolation
Synchronization

Keywords

  • Consistency
  • Diffusions
  • Nonsynchronous trading
  • Quadratic variation
  • Realized volatility

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Hayashi, T., & Yoshida, N. (2004). On covariance estimation for high-frequency financial data. In M. H. Hamza (Ed.), Proceedings of the Second IASTED International Conference on Financial Engineering and Applications (pp. 282-286). [437-801]

On covariance estimation for high-frequency financial data. / Hayashi, Takaki; Yoshida, Nakahiro.

Proceedings of the Second IASTED International Conference on Financial Engineering and Applications. ed. / M.H. Hamza. 2004. p. 282-286 437-801.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Hayashi, T & Yoshida, N 2004, On covariance estimation for high-frequency financial data. in MH Hamza (ed.), Proceedings of the Second IASTED International Conference on Financial Engineering and Applications., 437-801, pp. 282-286, Proceedings of the Second IASTED International Conference on Financial Engineering and Applications, Cambridge, MA, United States, 04/11/8.
Hayashi T, Yoshida N. On covariance estimation for high-frequency financial data. In Hamza MH, editor, Proceedings of the Second IASTED International Conference on Financial Engineering and Applications. 2004. p. 282-286. 437-801
Hayashi, Takaki ; Yoshida, Nakahiro. / On covariance estimation for high-frequency financial data. Proceedings of the Second IASTED International Conference on Financial Engineering and Applications. editor / M.H. Hamza. 2004. pp. 282-286
@inproceedings{f0feeb466dbc4fdcb70ea1f9d95ba610,
title = "On covariance estimation for high-frequency financial data",
abstract = "We consider the problem of estimating the covariance/correlation of two diffusion prices that are observed at discrete times in a nonsynchronous manner. The modem, popular approach in the literature, {"}realized{"} estimator, which is based on the sum of cross-products of intra-day log-price changes measured on regularly-spaced intervals over a day, is problematic because choice of regular interval size and data interpolation scheme may lead to unreliable estimation. We present a new estimation procedure recently proposed by [1], which is free of such {"}synchronization{"} of data, hence, free from biases or other problems caused by it. In particular, our estimators are shown to have consistency as the observation frequency (or the market liquidity) tends to infinity, which is not possessed by the realized estimators.",
keywords = "Consistency, Diffusions, Nonsynchronous trading, Quadratic variation, Realized volatility",
author = "Takaki Hayashi and Nakahiro Yoshida",
year = "2004",
language = "English",
isbn = "0889864179",
pages = "282--286",
editor = "M.H. Hamza",
booktitle = "Proceedings of the Second IASTED International Conference on Financial Engineering and Applications",

}

TY - GEN

T1 - On covariance estimation for high-frequency financial data

AU - Hayashi, Takaki

AU - Yoshida, Nakahiro

PY - 2004

Y1 - 2004

N2 - We consider the problem of estimating the covariance/correlation of two diffusion prices that are observed at discrete times in a nonsynchronous manner. The modem, popular approach in the literature, "realized" estimator, which is based on the sum of cross-products of intra-day log-price changes measured on regularly-spaced intervals over a day, is problematic because choice of regular interval size and data interpolation scheme may lead to unreliable estimation. We present a new estimation procedure recently proposed by [1], which is free of such "synchronization" of data, hence, free from biases or other problems caused by it. In particular, our estimators are shown to have consistency as the observation frequency (or the market liquidity) tends to infinity, which is not possessed by the realized estimators.

AB - We consider the problem of estimating the covariance/correlation of two diffusion prices that are observed at discrete times in a nonsynchronous manner. The modem, popular approach in the literature, "realized" estimator, which is based on the sum of cross-products of intra-day log-price changes measured on regularly-spaced intervals over a day, is problematic because choice of regular interval size and data interpolation scheme may lead to unreliable estimation. We present a new estimation procedure recently proposed by [1], which is free of such "synchronization" of data, hence, free from biases or other problems caused by it. In particular, our estimators are shown to have consistency as the observation frequency (or the market liquidity) tends to infinity, which is not possessed by the realized estimators.

KW - Consistency

KW - Diffusions

KW - Nonsynchronous trading

KW - Quadratic variation

KW - Realized volatility

UR - http://www.scopus.com/inward/record.url?scp=11144311841&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=11144311841&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:11144311841

SN - 0889864179

SN - 9780889864177

SP - 282

EP - 286

BT - Proceedings of the Second IASTED International Conference on Financial Engineering and Applications

A2 - Hamza, M.H.

ER -