The validity of bootstrap testing in the threshold framework

We consider bootstrap-based testing for threshold effects in non-linear threshold autoregressive (TAR) models. It is well-known that classic tests based on asymptotic theory tend to be oversized in the case of small, or even moderate sample sizes, or when the estimated parameters indicate non-stationarity, as often witnessed in the analysis of financial or climate data. To address the issue we propose a supremum Lagrange Multiplier test statistic (sLMb), where the null hypothesis specifies a linear autoregressive (AR) model against the alternative of a TAR model. We consider a recursive bootstrap applied to the sLMb statistic and establish its validity. This result is new, and requires the proof of non-standard results for bootstrap analysis in time series models; this includes a uniform bootstrap law of large numbers and a bootstrap functional central limit theorem. These new results can also be used as a general theoretical framework that can be adapted to other situations, such as regime-switching processes with exogenous threshold variables, or testing for structural breaks. The Monte Carlo evidence shows that the bootstrap test has correct empirical size even for small samples, and also no loss of empirical power when compared to the asymptotic test. Moreover, its performance is not affected if the order of the autoregression is estimated based on information criteria. Finally, we analyse a panel of short time series to assess the effect of warming on population dynamics.

[1]  Neville Davies,et al.  A portmanteau test for self-exciting threshold autoregressive-type nonlinearity in time series , 1986 .

[2]  Kung-Sik Chan,et al.  Testing for Threshold Effects in the TARMA Framework , 2021, Statistica Sinica.

[3]  J. Petruccelli,et al.  ON THE APPROXIMATION OF TIME SERIES BY THRESHOLD AUTOREGRESSIVE MODELS , 2016 .

[4]  K. Chan,et al.  Percentage Points of Likelihood Ratio Tests for Threshold Autoregression , 1991 .

[5]  Jonathan B. Hill.  WEAK-IDENTIFICATION ROBUST WILD BOOTSTRAP APPLIED TO A CONSISTENT MODEL SPECIFICATION TEST , 2018, Econometric Theory.

[6]  Bruce E. Hansen,et al.  Inference When a Nuisance Parameter Is Not Identified under the Null Hypothesis , 1996 .

[7]  Jirí Andel,et al.  On calculation of stationary density of autoregressive processes , 2000, Kybernetika.

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

[9]  K. Chan,et al.  Testing for threshold autoregression , 1990 .

[10]  H. Tong,et al.  Testing for threshold regulation in presence of measurement error with an application to the PPP hypothesis , 2020, 2002.09968.

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

[12]  R. Tsay.  Testing and modeling multivariate threshold models , 1998 .

[13]  Wai Keung Li,et al.  TESTING FOR DOUBLE THRESHOLD AUTOREGRESSIVE CONDITIONAL HETEROSCEDASTIC MODEL , 2000 .

[14]  Kung-Sik Chan,et al.  On Likelihood Ratio Tests for Threshold Autoregression , 1990 .

[15]  Kung-Sik Chan,et al.  Testing for Threshold Diffusion , 2017 .

[16]  H. Tong,et al.  Unit-root test within a threshold ARMA framework , 2020 .

[17]  Howell Tong,et al.  Threshold Models in Time Series Analysis-30 Years On , 2011 .

[18]  H. Tong.  Threshold models in time series analysis—Some reflections , 2015 .

[19]  Guodong Li,et al.  Testing a linear time series model against its threshold extension , 2011 .

[20]  W. K. Li,et al.  Testing for threshold autoregression with conditional heteroscedasticity , 1997 .

[21]  Common structure in panels of short ecological time-series , 2000, Proceedings of the Royal Society of London. Series B: Biological Sciences.

[22]  Greta Goracci.  An empirical study on the parsimony and descriptive power of TARMA models , 2020 .

[23]  Cheryl J. Briggs,et al.  What causes generation cycles in populations of stored-product moths? , 2000 .

[24]  D. Andrews.  Tests for Parameter Instability and Structural Change with Unknown Change Point , 1993 .

[25]  Howell Tong,et al.  TESTING FOR A LINEAR MA MODEL AGAINST THRESHOLD MA MODELS , 2005 .

[26]  A Primer on Bootstrap Testing of Hypotheses in Time Series Models: With an Application to Double Autoregressive Models , 2019, SSRN Electronic Journal.

[27]  B. Hansen.  Threshold autoregression in economics , 2011 .

[28]  Greta Goracci.  Revisiting the Canadian Lynx Time Series Analysis Through TARMA Models , 2020 .

[29]  Timo Teräsvirta,et al.  Testing linearity against smooth transition autoregressive models , 1988 .

[30]  Kung-Sik Chan,et al.  Guest Editors’ Introduction: Regime Switching and Threshold Models , 2017 .

[31]  H. Tong,et al.  From patterns to processes: phase and density dependencies in the Canadian lynx cycle. , 1998, Proceedings of the National Academy of Sciences of the United States of America.

[32]  On the Consistency of Bootstrap Testing for a Parameter on the Boundary of the Parameter Space , 2017 .

[33]  R. Knell,et al.  Warming at the population level: Effects on age structure, density, and generation cycles , 2019, Ecology and evolution.

[34]  H. Tong.  Non-linear time series. A dynamical system approach , 1990 .

[35]  D. Andrews.  Tests for Parameter Instability and Structural Change with Unknown Change Point , 1993 .