Changepoint Estimation in a Segmented Linear Regression via Empirical Likelihood

For a segmented regression system with an unknown changepoint over two domains of a predictor, a new empirical likelihood ratio statistic is proposed to test the null hypothesis of no change. Under the null hypothesis of no change, the proposed test statistic is shown empirically to be Gumbel distributed with robust location and scale estimators against various parameter settings and error distributions. A power analysis is conducted to illustrate the performance of the test. Under the alternative hypothesis with a changepoint, the test statistic is utilized to estimate the changepoint between the two domains. A comparison of the frequency distributions between the proposed estimator and two parametric methods indicates that the proposed method is effective in capturing the true changepoint.

[1]  Jie Chen.  Testing for a change point in linear regression models , 1998 .

[2]  Edit Gombay,et al.  An application of the maximum likelihood test to the change-point problem , 1994 .

[3]  Art B. Owen,et al.  Empirical Likelihood for Linear Models , 1991 .

[4]  P. Feder.  The Log Likelihood Ratio in Segmented Regression , 1975 .

[5]  K Ulm,et al.  A statistical method for assessing a threshold in epidemiological studies. , 1991, Statistics in medicine.

[6]  Lauren Bin Dong,et al.  Department of Economics Econometrics Working Paper Ewp0405 Tresting for Structural Change in Regression: an Empirical Likelihood Approach , 2022 .

[7]  G. Chow.  Tests of equality between sets of coefficients in two linear regressions (econometrics voi 28 , 1960 .

[8]  R. Davies.  Hypothesis Testing when a Nuisance Parameter is Present Only Under the Alternatives , 1987 .

[9]  P. K. Bhattacharya,et al.  Some aspects of change-point analysis , 1994 .

[10]  R. Davies.  Hypothesis testing when a nuisance parameter is present only under the alternative , 1977 .

[11]  D. Hinkley.  Inference about the intersection in two-phase regression , 1969 .

[12]  L. Horváth,et al.  Limit Theorems in Change-Point Analysis , 1997 .

[13]  Achim Zeileis,et al.  Implementing a class of structural change tests: An econometric computing approach , 2006, Comput. Stat. Data Anal..

[14]  A. John Bailer,et al.  Statistics for Environmental Biology and Toxicology , 2020 .

[15]  Narayanaswamy Balakrishnan,et al.  Asymptotic Methods in Probability and Statistics with Applications , 2012 .

[16]  Lieven Verschaffel,et al.  Using segmented linear regression models with unknown change points to analyze strategy shifts in cognitive tasks , 2001, Behavior research methods, instruments, & computers : a journal of the Psychonomic Society, Inc.

[17]  Soyoung Ryu,et al.  Estimating tree resin dose effect on termites , 2006 .

[18]  I. Fiteni,et al.  τ-estimators of regression models with structural change of unknown location , 2004 .

[19]  A. F. Smith,et al.  Straight Lines with a Change‐Point: A Bayesian Analysis of Some Renal Transplant Data , 1980 .

[20]  E. Vieth.  Fitting piecewise linear regression functions to biological responses. , 1989, Journal of applied physiology.

[21]  R. Quandt.  The Estimation of the Parameters of a Linear Regression System Obeying Two Separate Regimes , 1958 .

[22]  Xiaotong Shen,et al.  Empirical Likelihood , 2002 .

[23]  D. Hinkley.  Inference in Two-Phase Regression , 1971 .

[24]  Douglas M. Hawkins.  A Note on Continuous and Discontinuous Segmented Regressions , 1980 .

[25]  Andrej Pázman,et al.  Nonlinear Regression , 2019, Handbook of Regression Analysis With Applications in R.

[26]  Hans-Peter Piepho,et al.  Inference for the Break Point in Segmented Regression with Application to Longitudinal Data , 2003 .

[27]  E. Gbur,et al.  Use of Segmented Regression in the Determination of the Base Temperature in Heat Accumulation Models 1 , 1979 .

[28]  R. Davies.  Hypothesis testing when a nuisance parameter is present only under the alternative , 1977 .

[29]  Robert Lund,et al.  Detection of Undocumented Changepoints: A Revision of the Two-Phase Regression Model , 2002 .

[30]  V. Muggeo.  Estimating regression models with unknown break‐points , 2003, Statistics in medicine.

[31]  M. Lesperance,et al.  PIECEWISE REGRESSION: A TOOL FOR IDENTIFYING ECOLOGICAL THRESHOLDS , 2003 .

[32]  E. Guallar,et al.  Use of two-segmented logistic regression to estimate change-points in epidemiologic studies. , 1998, American journal of epidemiology.

[33]  Heping Zhang,et al.  Confidence regions in broken line regression , 1994 .

[34]  R. Quandt.  Tests of the Hypothesis That a Linear Regression System Obeys Two Separate Regimes , 1960 .

[35]  Hira L. Koul,et al.  Asymptotics of maximum likelihood estimator in a two-phase linear regression model , 2002 .

[36]  G. Seber,et al.  Nonlinear Regression: Seber/Nonlinear Regression , 2005 .

[37]  P. Feder.  On Asymptotic Distribution Theory in Segmented Regression Problems-- Identified Case , 1975 .

[38]  I. Johnstone,et al.  On Hotelling's Formula for the Volume of Tubes and Naiman's Inequality , 1989 .

[39]  Robert Lund,et al.  Mean shift testing in correlated data , 2011 .

[40]  Lianfen Qian,et al.  On maximum likelihood estimators for a threshold autoregression 1 1 Research was partly supported by , 1998 .