On long-run covariance matrix estimation with the truncated flat kernel

Chang Ching Lin; Shinichi Sataka

doi:10.1007/978-1-4614-1653-1_15

On long-run covariance matrix estimation with the truncated flat kernel

Chang Ching Lin, Shinichi Sataka

Department of Economics

Research output: Chapter in Book/Report/Conference proceeding › Chapter

Abstract

Despite its large sample efficiency, the truncated flat kernel (TF) estimator of long-run covariance matrices is seldom used, because it occasionally gives a nonpositive semidefinite estimate and sometimes performs poorly in small samples, compared to other familiar kernel estimators. This paper proposes simple modifications to the TF estimator to enforce the positive definiteness without sacrificing the large sample efficiency and make the estimator more reliable in small samples through better utilization of the bias-variance trade-off. We study the large sample properties of the modified TF estimators and verify their improved small-sample performances by Monte Carlo simulations.

Original language	English
Title of host publication	Recent Advances and Future Directions in Causality, Prediction, and Specification Analysis
Subtitle of host publication	Essays in Honor of Halbert L. White Jr
Publisher	Springer New York
Pages	383-410
Number of pages	28
ISBN (Electronic)	9781461416531
ISBN (Print)	9781461416524
DOIs	https://doi.org/10.1007/978-1-4614-1653-1_15
Publication status	Published - 2013 Jan 1

All Science Journal Classification (ASJC) codes

General Economics,Econometrics and Finance
General Business,Management and Accounting

Access to Document

10.1007/978-1-4614-1653-1_15

Cite this

@inbook{b813f98cdc044500bc7be3b082ac6f9d,

title = "On long-run covariance matrix estimation with the truncated flat kernel",

abstract = "Despite its large sample efficiency, the truncated flat kernel (TF) estimator of long-run covariance matrices is seldom used, because it occasionally gives a nonpositive semidefinite estimate and sometimes performs poorly in small samples, compared to other familiar kernel estimators. This paper proposes simple modifications to the TF estimator to enforce the positive definiteness without sacrificing the large sample efficiency and make the estimator more reliable in small samples through better utilization of the bias-variance trade-off. We study the large sample properties of the modified TF estimators and verify their improved small-sample performances by Monte Carlo simulations.",

author = "Lin, {Chang Ching} and Shinichi Sataka",

note = "Publisher Copyright: {\textcopyright} Springer Science+Business Media New York 2013.",

year = "2013",

month = jan,

day = "1",

doi = "10.1007/978-1-4614-1653-1_15",

language = "English",

isbn = "9781461416524",

pages = "383--410",

booktitle = "Recent Advances and Future Directions in Causality, Prediction, and Specification Analysis",

publisher = "Springer New York",

address = "United States",

}

TY - CHAP

T1 - On long-run covariance matrix estimation with the truncated flat kernel

AU - Lin, Chang Ching

AU - Sataka, Shinichi

PY - 2013/1/1

Y1 - 2013/1/1

N2 - Despite its large sample efficiency, the truncated flat kernel (TF) estimator of long-run covariance matrices is seldom used, because it occasionally gives a nonpositive semidefinite estimate and sometimes performs poorly in small samples, compared to other familiar kernel estimators. This paper proposes simple modifications to the TF estimator to enforce the positive definiteness without sacrificing the large sample efficiency and make the estimator more reliable in small samples through better utilization of the bias-variance trade-off. We study the large sample properties of the modified TF estimators and verify their improved small-sample performances by Monte Carlo simulations.

AB - Despite its large sample efficiency, the truncated flat kernel (TF) estimator of long-run covariance matrices is seldom used, because it occasionally gives a nonpositive semidefinite estimate and sometimes performs poorly in small samples, compared to other familiar kernel estimators. This paper proposes simple modifications to the TF estimator to enforce the positive definiteness without sacrificing the large sample efficiency and make the estimator more reliable in small samples through better utilization of the bias-variance trade-off. We study the large sample properties of the modified TF estimators and verify their improved small-sample performances by Monte Carlo simulations.

UR - http://www.scopus.com/inward/record.url?scp=84948969690&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84948969690&partnerID=8YFLogxK

U2 - 10.1007/978-1-4614-1653-1_15

DO - 10.1007/978-1-4614-1653-1_15

M3 - Chapter

AN - SCOPUS:84948969690

SN - 9781461416524

SP - 383

EP - 410

BT - Recent Advances and Future Directions in Causality, Prediction, and Specification Analysis

PB - Springer New York

ER -

On long-run covariance matrix estimation with the truncated flat kernel

Abstract

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this