GMM and QML asymptotic standard deviations in stochastic volatility models: Comments on Ruiz (1994)

Torben Andersen; Bent E. Sørensen

doi:10.1016/0304-4076(95)01799-2

GMM and QML asymptotic standard deviations in stochastic volatility models: Comments on Ruiz (1994)

Torben Andersen^*, Bent E. Sørensen

^*Corresponding author for this work

Finance

Research output: Contribution to journal › Article › peer-review

13 Scopus citations

Abstract

This note describes a practical procedure for arbitrarily precise calculation of the GMM asymptotic standard deviations for the parameters in a stochastic volatility model. Earlier results provided by Ruiz (1994) are flawed on this point. The correct numbers are in some cases orders of magnitude different from the prior published figures. The implications regarding the relative efficiency between GMM and QML estimates are briefly discussed.

Original language	English (US)
Pages (from-to)	397-403
Number of pages	7
Journal	Journal of Econometrics
Volume	76
Issue number	1-2
DOIs	https://doi.org/10.1016/0304-4076(95)01799-2
State	Published - Jan 1 1997

Keywords

Asymptotic bias
Optimal weighting matrix
Relative efficiency

ASJC Scopus subject areas

Economics and Econometrics
Applied Mathematics

Access to Document

10.1016/0304-4076(95)01799-2

Cite this

@article{9fe81839cdfc494a8707b90731815395,

title = "GMM and QML asymptotic standard deviations in stochastic volatility models: Comments on Ruiz (1994)",

abstract = "This note describes a practical procedure for arbitrarily precise calculation of the GMM asymptotic standard deviations for the parameters in a stochastic volatility model. Earlier results provided by Ruiz (1994) are flawed on this point. The correct numbers are in some cases orders of magnitude different from the prior published figures. The implications regarding the relative efficiency between GMM and QML estimates are briefly discussed.",

keywords = "Asymptotic bias, Optimal weighting matrix, Relative efficiency",

author = "Torben Andersen and S{\o}rensen, {Bent E.}",

year = "1997",

month = jan,

day = "1",

doi = "10.1016/0304-4076(95)01799-2",

language = "English (US)",

volume = "76",

pages = "397--403",

journal = "Journal of Econometrics",

issn = "0304-4076",

publisher = "Elsevier BV",

number = "1-2",

}

TY - JOUR

T1 - GMM and QML asymptotic standard deviations in stochastic volatility models

T2 - Comments on Ruiz (1994)

AU - Andersen, Torben

AU - Sørensen, Bent E.

PY - 1997/1/1

Y1 - 1997/1/1

N2 - This note describes a practical procedure for arbitrarily precise calculation of the GMM asymptotic standard deviations for the parameters in a stochastic volatility model. Earlier results provided by Ruiz (1994) are flawed on this point. The correct numbers are in some cases orders of magnitude different from the prior published figures. The implications regarding the relative efficiency between GMM and QML estimates are briefly discussed.

AB - This note describes a practical procedure for arbitrarily precise calculation of the GMM asymptotic standard deviations for the parameters in a stochastic volatility model. Earlier results provided by Ruiz (1994) are flawed on this point. The correct numbers are in some cases orders of magnitude different from the prior published figures. The implications regarding the relative efficiency between GMM and QML estimates are briefly discussed.

KW - Asymptotic bias

KW - Optimal weighting matrix

KW - Relative efficiency

UR - http://www.scopus.com/inward/record.url?scp=0040034892&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0040034892&partnerID=8YFLogxK

U2 - 10.1016/0304-4076(95)01799-2

DO - 10.1016/0304-4076(95)01799-2

M3 - Article

AN - SCOPUS:0040034892

VL - 76

SP - 397

EP - 403

JO - Journal of Econometrics

JF - Journal of Econometrics

SN - 0304-4076

IS - 1-2

ER -

GMM and QML asymptotic standard deviations in stochastic volatility models: Comments on Ruiz (1994)

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this