Efficient robust estimation of time-series regression models

Research output: Contribution to journalArticleScientificpeer-review

Abstract

The paper studies a new class of robust regression estimators based on the two-step least weighted squares (2S-LWS) estimator which employs data-adaptive weights determined from the empirical distribution or quantile functions of regression residuals obtained from an initial robust fit. Just like many existing two-step robust methods, the proposed 2S-LWS estimator preserves robust properties of the initial robust estimate. However, contrary to the existing methods, the first-order asymptotic behavior of 2S-LWS is fully independent of the initial estimate under mild conditions. We propose data-adaptive weighting schemes that perform well both in the cross-section and time-series data and prove the asymptotic normality and efficiency of the resulting procedure. A simulation study documents these theoretical properties in finite samples.
Original languageEnglish
Pages (from-to)267-279
JournalApplications of Mathematics
Volume53
Issue number3
Publication statusPublished - 2008

Fingerprint

Robust Estimators
Efficient Estimation
Robust Estimation
Time Series Models
Time series
Regression Model
Robust Estimate
Robust Regression
Quantile Function
Empirical Distribution Function
Two-step Method
Asymptotic Efficiency
Regression Estimator
Weighted Least Squares
Robust Methods
Time Series Data
Asymptotic Normality
Weighting
Cross section
Regression

Cite this

@article{35a355e4d17b40bf8fdd83b402aa24fb,
title = "Efficient robust estimation of time-series regression models",
abstract = "The paper studies a new class of robust regression estimators based on the two-step least weighted squares (2S-LWS) estimator which employs data-adaptive weights determined from the empirical distribution or quantile functions of regression residuals obtained from an initial robust fit. Just like many existing two-step robust methods, the proposed 2S-LWS estimator preserves robust properties of the initial robust estimate. However, contrary to the existing methods, the first-order asymptotic behavior of 2S-LWS is fully independent of the initial estimate under mild conditions. We propose data-adaptive weighting schemes that perform well both in the cross-section and time-series data and prove the asymptotic normality and efficiency of the resulting procedure. A simulation study documents these theoretical properties in finite samples.",
author = "P. Cizek",
note = "Appeared previously as CentER DP 2007-95",
year = "2008",
language = "English",
volume = "53",
pages = "267--279",
journal = "Applications of Mathematics",
issn = "0862-7940",
publisher = "Springer Netherlands",
number = "3",

}

Efficient robust estimation of time-series regression models. / Cizek, P.

In: Applications of Mathematics, Vol. 53, No. 3, 2008, p. 267-279.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Efficient robust estimation of time-series regression models

AU - Cizek, P.

N1 - Appeared previously as CentER DP 2007-95

PY - 2008

Y1 - 2008

N2 - The paper studies a new class of robust regression estimators based on the two-step least weighted squares (2S-LWS) estimator which employs data-adaptive weights determined from the empirical distribution or quantile functions of regression residuals obtained from an initial robust fit. Just like many existing two-step robust methods, the proposed 2S-LWS estimator preserves robust properties of the initial robust estimate. However, contrary to the existing methods, the first-order asymptotic behavior of 2S-LWS is fully independent of the initial estimate under mild conditions. We propose data-adaptive weighting schemes that perform well both in the cross-section and time-series data and prove the asymptotic normality and efficiency of the resulting procedure. A simulation study documents these theoretical properties in finite samples.

AB - The paper studies a new class of robust regression estimators based on the two-step least weighted squares (2S-LWS) estimator which employs data-adaptive weights determined from the empirical distribution or quantile functions of regression residuals obtained from an initial robust fit. Just like many existing two-step robust methods, the proposed 2S-LWS estimator preserves robust properties of the initial robust estimate. However, contrary to the existing methods, the first-order asymptotic behavior of 2S-LWS is fully independent of the initial estimate under mild conditions. We propose data-adaptive weighting schemes that perform well both in the cross-section and time-series data and prove the asymptotic normality and efficiency of the resulting procedure. A simulation study documents these theoretical properties in finite samples.

M3 - Article

VL - 53

SP - 267

EP - 279

JO - Applications of Mathematics

JF - Applications of Mathematics

SN - 0862-7940

IS - 3

ER -