Abstract
Inferences from tests for non-stationarity depend critically on whether and how breaks and/or non-linearities are specified. Recent work has shown that wavelet transformations that separate a variable’s high and low frequency components can enhance the performance of unit root and stationarity tests. This note provides response surface estimates of finite sample, lag-adjusted critical values and approximate probability values for an Augmented Dickey–Fuller type wavelet test that includes a Fourier term allowing for smooth breaks in the series. Applications highlight the practical benefits.
Similar content being viewed by others
Data availability
Matlab programs to replicate the analysis and calculate finite sample lag adjusted critical values and probability values are available as electronic supplementary material and also at http://web.business.queensu.ca/faculty/psephton under MatlabFiles (FWADF.zip) and on request to the author (Peter.Sephton@queensu.ca).
Notes
An entire issue of Econometrics (2017) was devoted to unit roots and structural breaks—see Perron (2017).
Aydin and Pata (2020) show the Haar wavelet transform led to the greatest test power and is employed here.
For example, the FWADF constant and trend test adds two regressors to the testing equation, reducing the degrees of freedom, which in small samples, could affect the distribution of the test statistic. Cook (2001) showed that using lag-adjusted critical values can increase the power of unit root tests.
Experimentation led to g set to 3 and h set to 2.5. Fractional powers of the lag adjusted regressors increased the explanatory power of the response surface regressions.
The wavelet test decomposes the series into low and high frequency components, using only half of the data contained in the low frequency component to estimate the test statistic. The sample size associated with the length of the low frequency component was employed as the “driver” of the error variance.
Variations in the goodness of fit across quantiles is not unusual: Kripfganz and Schneider (2020) report adjusted R-squared values as low as 0.77 for some response surfaces.
Here the t distribution was employed to calculate the approximate probability values. Kripfganz and Schneider (2020) followed a similar approach.
Tests that include only a constant and the Fourier term did not reject the null of a unit root for all series at five percent with both the AIC and BIC (with the exception Brazilian Naturals and the BIC). The coefficients on the trend terms were significantly different from zero at five percent.
References
Aydin, M., & Pata, U. (2020). Are shocks to disaggregated renewable energy consumption permanent or temporary for the USA? Wavelet based unit root test with smooth structural shifts. Energy, 207, 118245. https://doi.org/10.1016/j.energy.2020.118245
Bastianin, A., Lanza, A., & Manera, M. (2018). Economic impacts of El Niño Southern oscillation: Evidence from the Columbian coffee market. Agricultural Economics, 49(5), 623–633. https://doi.org/10.1111/agec.12447
Cheung, Y., & Lai, K. (1995). Lag order and critical values of a modified Dickey–Fuller test. Oxford Bulletin of Economics and Statistics, 57, 411–419.
Cook, S. (2001). Finite sample critical values of the augmented Dickey-Fuller statistic: ANote on lag order. Economic Issues, 16(2), 31–38.
Dickey, D., & Fuller, W. (1979). Distribution of the estimators for autoregressive time series with a unit root. Journal of the American Statistical Association, 74(366a), 427–431.
Elliot, G., Rothenberg, T., & Stock, J. (1996). Efficient tests for an autoregressive unit root. Econometrica, 64, 813–836.
Enders, W., & Lee, J. (2012). The Flexible Fourier Form and Dickey-Fuller Unit Root Tests. Economics Letters, 117(1), 196–199.
Erdogan, S., & Solarin, S. (2021). Stochastic convergence in carbon emissions based on a new fourier-based wavelet unit root test. Environmental Science and Pollution Research, 28, 1887–21899. https://doi.org/10.1007/s11356-020-12033-y
Ergolu, B., & Soybilgen, B. (2018). On the performance of wavelet based unit root tests. Journal of Risk and Financial Management, 11(47), 47.
Hobijn, B., Franses, P., & Ooms, M. (2004). Generalizations of the KPSS-test for stationarity. Statistica Neerlandica, 58(4), 483–502.
Kripfganz, S., & Schneider, D. (2020). Response surface regressions for critical value bounds and approximate p-values in equilibrium correction models. Oxford Bulletin of Economics and Statistics, 82(6), 1456–1481.
Melo-Velandia, L., Orozco-Vanegas, C., & Parra-Amado, D. (2022). Extreme weather events and high columbian food prices: A non-stationary extreme value approach. Borradores De Economica, 1189, 1–15.
Omay, T. (2015). Fractional frequency flexible fourier form to approximate smooth breaks in unit root testing. Economics Letters, 134, 123–126.
Perron, P. (2017). Unit roots and structural breaks. Econometrics, 5(22), 1–3.
Sephton, P. (1995). Response surface estimates of the KPSS stationsarity test. Economics Letters, 47, 255–261.
Sephton, P. (2008). Critical values of the augmented fractional Dickey Fuller test. Empirical Economics, 35, 437–450.
Sephton, P. (2009). critical values for the augmented efficient wald test for fractional unit roots. Empirical Economics, 37, 615–626.
Sephton, P. (2017). Finite sample critical values of the generalized KPSS stationarity test. Computational Economics, 50, 161–172.
Sephton, P. (2019). El Niño, La Niña, and a Cup of Joe. Energy Economics, 84, 104503. https://doi.org/10.1016/j.eneco.2019.104503
Sephton, P. (2022). Finite sample lag adjusted critical values of the ADF-GLS test. Computational Economics, 59, 177–183.
Shahbaz, M., & Sinha, A. (2019). Environmental kuznets curve for CO2 emissions: A literature survey. Journal of Economic Studies, 46(1), 106–168.
Solarin, S., Erdogan, S., & Bello, M. (2022). Convergence analysis of ammonia emissions by sector and fuel source in OECD countries from 1750 to 2019 using a new fourier-centric wavelet approach. Evironmental Scence and Pollution Research. https://doi.org/10.1007/s11356-022-21007-1
Ubilava, D. (2012). El Niño, La Niña and World coffee price dynamics. Agricultural Economics, 43, 17–26.
Acknowledgements
I thank Mucahit Aydin and Tolga Omay for helpful comments.
Funding
No funds grants or other support were received during the preparation of this manuscript.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Competing interests
The author declares he has no financial or non-financial interests to disclose.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Sephton, P.S. Finite Sample Lag Adjusted Critical Values and Probability Values for the Fourier Wavelet Unit Root Test. Comput Econ (2023). https://doi.org/10.1007/s10614-023-10458-4
Accepted:
Published:
DOI: https://doi.org/10.1007/s10614-023-10458-4