Abstract:Both seasonal unit roots and periodic variation can be prevalent in seasonal data. When testing seasonal unit roots under periodic variation, the validity of the existing methods, such as the HEGY test, remains unknown. This paper analyzes the behavior of the augmented HEGY test and the unaugmented HEGY test under periodic variation. It turns out that the asymptotic null distributions of the HEGY statistics testing the single roots at $1$ or $-1$ when there is periodic variation are identical to the asymptotic null distributions when there is no periodic variation. On the other hand, the asymptotic null distributions of the statistics testing any coexistence of roots at $1$, $-1$, $i$, or $-i$ when there is periodic variation are non-standard and are different from the asymptotic null distributions when there is no periodic variation. Therefore, when periodic variation exists, HEGY tests are not directly applicable to the joint tests for any concurrence of seasonal unit roots. As a remedy, bootstrap is proposed; in particular, the augmented HEGY test with seasonal independent and identically distributed (iid) bootstrap and the unaugmented HEGY test with seasonal block bootstrap are implemented. The consistency of these bootstrap procedures is established. The finite-sample behavior of these bootstrap tests is illustrated via simulation and prevails over their competitors'. Finally, these bootstrap tests are applied to detect the seasonal unit roots in various economic time series.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is the validity of existing seasonal unit root test methods (such as the HEGY test) in the presence of cyclical changes. Specifically, when the data has cyclical change characteristics, whether the existing test methods are still valid and how to design new test methods to meet this challenge. ### Background and Problem Description In time - series analysis, seasonal unit roots and cyclical changes are two common phenomena. Seasonal unit roots refer to the non - stationary random seasonality exhibited by a time series within a specific season, while cyclical changes refer to the variation of the parameters of the time - series generation process with the change of seasons. These two phenomena may co - exist in seasonal data. For example, the seasonal consumption data in the UK shows both cyclical changes and seasonal unit roots. However, the validity of existing seasonal unit root test methods (such as the HEGY test) in dealing with data with cyclical changes has not been fully verified. In particular, when the data has cyclical changes, the asymptotic distribution properties of the HEGY test may change, leading to inaccurate test results. ### Main Contributions of the Paper 1. **Analysis of the Validity of Existing Methods**: - The asymptotic distribution properties of the augmented HEGY test and the non - augmented HEGY test in the presence of cyclical changes are studied. - It is found that when testing a single root (1 or - 1), the asymptotic distributions of the augmented HEGY test and the non - augmented HEGY test are the same as those in the absence of cyclical changes. - When testing the simultaneous existence of multiple roots (1, - 1, i, - i), the asymptotic distribution is non - standard and different from that in the absence of cyclical changes. 2. **Proposing New Test Methods**: - To meet the challenges brought by cyclical changes, two new test methods based on the bootstrap are proposed: - The seasonally independent and identically distributed (iid) bootstrap method for the augmented HEGY test. - The seasonal block bootstrap method for the non - augmented HEGY test. - These new methods improve the accuracy of the test by simulating the cyclical structure in the data generation process. 3. **Theoretical and Empirical Analysis**: - The consistency of these bootstrap methods is proved. - Through simulation experiments, it is verified that the performance of these bootstrap methods in finite samples is better than that of existing methods. - These new methods are applied to actual economic time - series data, demonstrating their effectiveness in practical applications. ### Conclusion This paper, through in - depth analysis of the behavior of seasonal unit root test methods in the presence of cyclical changes, discovers the limitations of existing methods and proposes new test methods based on the bootstrap. These new methods are not only strictly proven theoretically but also show better performance in practical applications. This provides an effective tool for dealing with seasonal time - series data with cyclical change characteristics.

Bootstrap Seasonal Unit Root Test under Periodic Variation

Linear Process Bootstrap Unit Root Test

Specification Testing in Nonstationary Time Series Models

The Grid Bootstrap for Continuous Time Models

The Limit Distribution of the Bootstrap for the Unit Root Test Statistic when the Residuals Are Dependent

Bootstrapping non-stationary stochastic volatility

Testing for a Unit Root with Nonstationary Nonlinear Heteroskedasticity

Specification Tests for Time-Varying Coefficient Models

Bootstrap Adaptive Lasso Solution Path Unit Root Tests

Diagnostic Checking for the Adequacy of Nonlinear Time Series Models

Adaptive LAD-Based Bootstrap Unit Root Tests under Unconditional Heteroskedasticity

An Introduction to Bootstrap Theory in Time Series Econometrics

A General Approach to Testing Volatility Models in Time Series

The validity of bootstrap testing in the threshold framework

Unit Roots in Time Series with Changepoints

Generalized Spectral Tests for Conditional Mean Models in Time Series with Conditional Heteroscedasticity of Unknown Form

Unit Root Test in Panel Data Basing on the Limiting Spectral Distribution of Large-Dimensional Random Matrix

Testing Equality of Autocovariance Operators for Functional Time Series

Bootstrap Inference in Cointegrating Regressions: Traditional and Self-Normalized Test Statistics

A Global Wavelet Based Bootstrapped Test of Covariance Stationarity

Approach to Testing Nonlinear Time Series Models