Positive semidefinite integrated covariance estimation, factorizations and asynchronicity

Kris Boudt, Sébastien Laurent, Asger Lunde, Rogier Quaedvlieg, Sauri Arregui Orimar

Research output: Contribution to journalJournal articleResearchpeer-review

4 Citations (Scopus)

Abstract

An estimator of the ex-post covariation of log-prices under asynchronicity and microstructure noise is proposed. It uses the Cholesky factorization of the covariance matrix in order to exploit the heterogeneity in trading intensities to estimate the different parameters sequentially with as many observations as possible. The estimator is positive semidefinite by construction. We derive asymptotic results and confirm their good finite sample properties by means of a Monte Carlo simulation. In the application we forecast portfolio Value-at-Risk and sector risk exposures for a portfolio of 52 stocks. We find that the dynamic models utilizing the proposed high-frequency estimator provide statistically and economically superior forecasts.
Original languageEnglish
JournalJournal of Econometrics
Volume196
Issue number2
Pages (from-to)347-367
Number of pages21
ISSN0304-4076
DOIs
Publication statusPublished - 2017

Fingerprint

Integrated
Estimator
Microstructure noise
Trade intensity
Monte Carlo simulation
Finite sample properties
Risk exposure
Covariance matrix
Value at risk

Cite this

Boudt, Kris ; Laurent, Sébastien ; Lunde, Asger ; Quaedvlieg, Rogier ; Orimar, Sauri Arregui. / Positive semidefinite integrated covariance estimation, factorizations and asynchronicity. In: Journal of Econometrics. 2017 ; Vol. 196, No. 2. pp. 347-367.
@article{4d7fd03082714904939ee671fa1a1fce,
title = "Positive semidefinite integrated covariance estimation, factorizations and asynchronicity",
abstract = "An estimator of the ex-post covariation of log-prices under asynchronicity and microstructure noise is proposed. It uses the Cholesky factorization of the covariance matrix in order to exploit the heterogeneity in trading intensities to estimate the different parameters sequentially with as many observations as possible. The estimator is positive semidefinite by construction. We derive asymptotic results and confirm their good finite sample properties by means of a Monte Carlo simulation. In the application we forecast portfolio Value-at-Risk and sector risk exposures for a portfolio of 52 stocks. We find that the dynamic models utilizing the proposed high-frequency estimator provide statistically and economically superior forecasts.",
author = "Kris Boudt and S{\'e}bastien Laurent and Asger Lunde and Rogier Quaedvlieg and Orimar, {Sauri Arregui}",
year = "2017",
doi = "10.1016/j.jeconom.2016.09.016",
language = "English",
volume = "196",
pages = "347--367",
journal = "Journal of Econometrics",
issn = "0304-4076",
publisher = "Elsevier",
number = "2",

}

Positive semidefinite integrated covariance estimation, factorizations and asynchronicity. / Boudt, Kris; Laurent, Sébastien; Lunde, Asger; Quaedvlieg, Rogier; Orimar, Sauri Arregui.

In: Journal of Econometrics, Vol. 196, No. 2, 2017, p. 347-367.

Research output: Contribution to journalJournal articleResearchpeer-review

TY - JOUR

T1 - Positive semidefinite integrated covariance estimation, factorizations and asynchronicity

AU - Boudt, Kris

AU - Laurent, Sébastien

AU - Lunde, Asger

AU - Quaedvlieg, Rogier

AU - Orimar, Sauri Arregui

PY - 2017

Y1 - 2017

N2 - An estimator of the ex-post covariation of log-prices under asynchronicity and microstructure noise is proposed. It uses the Cholesky factorization of the covariance matrix in order to exploit the heterogeneity in trading intensities to estimate the different parameters sequentially with as many observations as possible. The estimator is positive semidefinite by construction. We derive asymptotic results and confirm their good finite sample properties by means of a Monte Carlo simulation. In the application we forecast portfolio Value-at-Risk and sector risk exposures for a portfolio of 52 stocks. We find that the dynamic models utilizing the proposed high-frequency estimator provide statistically and economically superior forecasts.

AB - An estimator of the ex-post covariation of log-prices under asynchronicity and microstructure noise is proposed. It uses the Cholesky factorization of the covariance matrix in order to exploit the heterogeneity in trading intensities to estimate the different parameters sequentially with as many observations as possible. The estimator is positive semidefinite by construction. We derive asymptotic results and confirm their good finite sample properties by means of a Monte Carlo simulation. In the application we forecast portfolio Value-at-Risk and sector risk exposures for a portfolio of 52 stocks. We find that the dynamic models utilizing the proposed high-frequency estimator provide statistically and economically superior forecasts.

U2 - 10.1016/j.jeconom.2016.09.016

DO - 10.1016/j.jeconom.2016.09.016

M3 - Journal article

VL - 196

SP - 347

EP - 367

JO - Journal of Econometrics

JF - Journal of Econometrics

SN - 0304-4076

IS - 2

ER -