Periodogram Based Unit Root Test
Synopsis
In this study authored by Mesut Balıbey, the necessity of stationarity in time series for producing accurate and reliable future forecasts is emphasized, and the application of periodogram-based unit root tests to eliminate non-stationarity is examined in detail. Using EViews 11 software, both the conventional Augmented Dickey-Fuller (ADF) test and the periodogram-based unit root test proposed by Akdi and Dickey (1998) were applied to annual wheat production data (log-production). The analysis results demonstrate that both methods are fully consistent with each other, showing that the raw wheat production series contains a unit root and is non-stationary, but becomes stationary once the first difference of the series is taken. Periodogram-based unit root tests offer significant methodological advantages over traditional approaches, such as requiring no model specification, being invariant to the mean, being free from sample size constraints, and being robust against seasonal fluctuations. Trend and seasonality components induced by natural disasters, economic crises, or epidemics distort the stationarity of time series, leading to inaccurate forecasting. Consequently, stabilizing the series through first-differencing is of vital importance for policymakers, agricultural producers, and researchers to execute effective strategic planning and derive dependable future predictions.
References
Akdi Y. and Dickey D.A. (1998). Periodograms of unit root time series: distributions and tests. Commun. Statist-Theory Meth. 27(1), 69-87.
Akdi Y. and Dickey D.A. (1999). Periodograms for seasonal time series with a unit root. Journal of The Turkish Statitical Association, Volume 2, Number 3, pp.153-164.
Akdi Y., Kalafatcılar K. M., Özcan K. M. (2010). Application of periodogram-based cointegration test for the analysis of the services and goods sector inflations. International Econometric Review (IER), vol. 2, issue 1, 3-10.
Akdi Y, Gölveren E, Okkaoğlu Y (2020a). Daily electrical energy consumption: periodicity, harmonic regression method and forecasting. Energy 191:116524
Akdi Y, Okkaoğlu Y, Gölveren E, Yücel ME (2020b) Estimation and forecasting of PM10 air pollution in Ankara via time series and harmonic regressions. Int J Environ Sci Technol 17:3677–3690. https://doi.org/10.1007/s13762-020-02705-0
Akdi, Y., Balıbey, M. ve Kayış, A. (2019). Petrol ve doğalgaz fiyatlarındaki periyodik dalgalanmalar. Fırat Üniversitesi Sosyal Bilimler Dergisi, 29 (2), 217-325. DOI:10.18069/firatsbed.559747
Balıbey, M , Akdi, Y . (2019). Döviz kurlarında periyodik dalgalanmalar. Atatürk Üniversitesi Sosyal Bilimler Enstitüsü Dergisi, 23 (1), 123-131. Retrieved from
https://dergipark.org.tr/en/pub/ataunisosbil/issue/43928/538466
Dickey, D.A., Hazsa, D.P. and Fuller, W.A. (1984). Testing for unit roots in seasonal time series, Journal of American Statistical Association, Volume 79, Number 386.
Tekin, K. ve Akdi, Y . (2014). Mevsimsel birim kök testleri: Türkiye sanayi üretim endeksi üzerine bir uygulama. Gazi Üniversitesi Sosyal Bilimler Dergisi, 1 (1). Retrieved from https://dergipark.org.tr/en/pub/gusbd/issue/7520/99116
Hylleberg, S., Engle, R.F., Granger, C.W.J. and Yoo, B.S. (1990). Seasoal integration and cointegration, Journal of Econometrics, 44,215-238.
Oyakhilomen, O. & Zibah, R. G. (2014). Agricultural production and economic growth in Nigeria: Implication for rural poverty alleviation. Quarterly Journal of International Agriculture, 53(892-2016-65234), 207-223.
