Detecting At-Most-m-Changes in Linear Regression Models

Lajos Horvath; William Pouliot; Shixuan Wang

doi:10.1111/jtsa.12228

Detecting At-Most-m-Changes in Linear Regression Models

Lajos Horvath, William Pouliot, Shixuan Wang

Economics

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

4 Citations (Scopus)

166 Downloads (Pure)

Abstract

In this paper we provide a new procedure to test for at most m changes in the
time–dependent regression model Our procedure is based on weighted sums of the residuals, incorporating the possibility of m changes. The weak limit of the proposed test statistic is the sum of two double exponential random variables. A small Monte Carlo simulation illustrates the applicability of the limit results in case of small and moderate sample sizes. We compare the new method to the CUSUM and standardized (weighted) CUSUM procedures and obtain
the power curves of the test statistics under the alternative. We apply our method to find changes in the unconditional four factor CAPM.

Original language	English
Pages (from-to)	552-590
Number of pages	39
Journal	Journal of Time Series Analysis
Volume	38
Issue number	4
Early online date	26 Dec 2016
DOIs	https://doi.org/10.1111/jtsa.12228
Publication status	Published - Jul 2017

Keywords

Change point
Bernoulli shifts
weak approximation
weighted CUSUM
residuals
linear regression models

Access to Document

10.1111/jtsa.12228Licence: None: All rights reserved

2C-jtsa-NOV12_2016
This is the peer reviewed version of the following article: Horváth, L., Pouliot, W., and Wang, S. (2016) Detecting at-Most-m Changes in Linear Regression Models. J. Time Ser. Anal., doi: 10.1111/jtsa.12228, which has been published in final form at 10.1111/jtsa.12228. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Self-Archiving.
Accepted author manuscript, 938 KBLicence: Other (please specify with Rights Statement)

Cite this

@article{ac1f59e461fb4ca3ad3a56447d9fe480,

title = "Detecting At-Most-m-Changes in Linear Regression Models",

abstract = "In this paper we provide a new procedure to test for at most m changes in thetime–dependent regression model Our procedure is based on weighted sums of the residuals, incorporating the possibility of m changes. The weak limit of the proposed test statistic is the sum of two double exponential random variables. A small Monte Carlo simulation illustrates the applicability of the limit results in case of small and moderate sample sizes. We compare the new method to the CUSUM and standardized (weighted) CUSUM procedures and obtainthe power curves of the test statistics under the alternative. We apply our method to find changes in the unconditional four factor CAPM.",

keywords = "Change point, Bernoulli shifts, weak approximation, weighted CUSUM, residuals, linear regression models",

author = "Lajos Horvath and William Pouliot and Shixuan Wang",

year = "2017",

month = jul,

doi = "10.1111/jtsa.12228",

language = "English",

volume = "38",

pages = "552--590",

journal = "Journal of Time Series Analysis",

issn = "0143-9782",

publisher = "Wiley",

number = "4",

}

TY - JOUR

T1 - Detecting At-Most-m-Changes in Linear Regression Models

AU - Horvath, Lajos

AU - Pouliot, William

AU - Wang, Shixuan

PY - 2017/7

Y1 - 2017/7

N2 - In this paper we provide a new procedure to test for at most m changes in thetime–dependent regression model Our procedure is based on weighted sums of the residuals, incorporating the possibility of m changes. The weak limit of the proposed test statistic is the sum of two double exponential random variables. A small Monte Carlo simulation illustrates the applicability of the limit results in case of small and moderate sample sizes. We compare the new method to the CUSUM and standardized (weighted) CUSUM procedures and obtainthe power curves of the test statistics under the alternative. We apply our method to find changes in the unconditional four factor CAPM.

AB - In this paper we provide a new procedure to test for at most m changes in thetime–dependent regression model Our procedure is based on weighted sums of the residuals, incorporating the possibility of m changes. The weak limit of the proposed test statistic is the sum of two double exponential random variables. A small Monte Carlo simulation illustrates the applicability of the limit results in case of small and moderate sample sizes. We compare the new method to the CUSUM and standardized (weighted) CUSUM procedures and obtainthe power curves of the test statistics under the alternative. We apply our method to find changes in the unconditional four factor CAPM.

KW - Change point

KW - Bernoulli shifts

KW - weak approximation

KW - weighted CUSUM

KW - residuals

KW - linear regression models

U2 - 10.1111/jtsa.12228

DO - 10.1111/jtsa.12228

M3 - Article

SN - 0143-9782

VL - 38

SP - 552

EP - 590

JO - Journal of Time Series Analysis

JF - Journal of Time Series Analysis

IS - 4

ER -

Detecting At-Most-m-Changes in Linear Regression Models

Abstract

Keywords

Access to Document

Fingerprint

Cite this