Spurious Trend in Stationary Series and its Implications

Article DOI: https://doi.org/10.57017/jaes.v15.3(69).12

Author(s):

Atiq-ur REHMAN Kashmir Institute of Economics, University of AJ&K, Pakistan
Ghulam Yahya KHAN Kashmir Institute of Economics, University of AJ&K, Pakistan
Muhammad SAIM HASHMI Faculty of Social Sciences and Humanities, Mirpur University of Science and Technology, Pakistan

Abstract:

Nelson and Kang (1984) showed that regression of a unit root time-series on a linear time trend provides significant results even if there is no forecast able association amongst the path of the time-series and linear trend. Using Monte Carlo simulations, this paper shows that phenomenon also exists in stationary time series and regression of a stationary time series on linear trend also produces significant results without the existence of any predictable relationship between the time -series and linear trend. The spurious trend is observable in most of the moderate sample sizes and sometimes in sufficiently large samples of size over 500 observations. The implications of these findings for unit root test procedures are discussed briefly.

How to cite:

Rehman, A., Khan, G.Y., Saim Hashmi, M. (2020). Spurious Trend in Stationary Series and its Implications. Journal of Applied Economic Sciences, Volume XV, Fall, 3(69), 636-646. https://doi.org/10.57017/jaes.v15.3(69).12

References:

[1] Agiakloglou, C. 2013. Resolving spurious regressions and serially correlated errors. Empirical Economics, 45(3): 1361-1366. https://doi.org/10.1007/s00181-012-0647-4.

[2] Agiakloglou, C, Tsimbos, C., and Tsimpanos, A. 2015. Is spurious behavior an issue for two independent stationary spatial autoregressive SAR (1) processes? Applied Economics Letters, 22(17): 1372-1377. https://doi.org/10.1080/13504851.2015.1031869

[3] Bacallado, S., Diaconis, P., and Holmes, S. 2015. De Finetti Priors using Markov chain Monte Carlo computations. Statistics and Computing, 25(4): 797-808. https://doi.org/10.1007/s11222-015-9562-9.

[4] Choi, I. 2013. Spurious fixed effects regression. Oxford Bulletin of Economics and Statistics, 75(2): 297-306. https://doi.org/10.1111/j.1468-0084.2011.00688.x

[5] Darné, O., and Diebold, C. 2002. A note on seasonal unit root tests. Quality and Quantity, 36(3): 305-310. https://doi.org/10.1023/A:1016032601197

[6] Deng, A. 2014. Understanding spurious regression in financial economics. Journal of Financial Econometrics, 12(1): 122–150. Winter. https://doi.org/10.1093/jjfinec/nbs025

[7] Dickey, D.A., and Fuller, W.A. 1979. Distribution of the estimators for autoregressive time series with a unit root. Journal of the American Statistical Association, 74(366a): 427−431. https://doi.org/10.1080/01621459.1979.10482531

[8] Diebold, C., and Giraud, V. 2005. A note on long memory time series. Quality and Quantity, 39(6): 827-836. https://doi.org/10.1007/s11135-004-0436-z

[9] Fernandez-Macho, J. 2015. Comment on testing for spurious and co integrated regressions: A wavelet approach. Journal of Applied Statistics, 42(8): 1759-1769. https://doi.org/10.1080/02664763.2015.1005583

[10] Granger, C.W.J., and Newbold, P. 1974. Spurious regressions in econometrics. Journal of Econometrics, 2(2): 111−120. https://doi.org/10.1016/0304-4076(74)90034-7

[11] Granger, C.W.J.IV., Namwon, H., and Yongil, J. 2001. Spurious regressions with stationary series. Applied Economics, 33(7): 899-904. https://doi.org/10.1080/00036840121734

[12] Harashima, T. 2020. The correlation between time preference and incomes is spurious: They are bridged by fluid intelligence. Journal of Applied Economic Sciences, Volume XV, Spring, 1(67): 106-123.

[13] Jentsch, C., Kreiss, J.P., Mantalos, P., and Paparoditis, E. 2012. Hybrid bootstrap aided unit root testing. Computational Statistics, 27(4): 779-797. https://doi.org/10.1007/s00180-011-0290-0

[14] Jin, H., Zhang, D, Zhang, J., and Zhang, S. 2017. Evidence of spurious regression driven by heavy-tailed observations with structural changes. Communications in Statistics-Simulation and Computation, 46(2): 1086-1101. https://doi.org/10.1080/03610918.2014.990098

[15] Jin, H, Zhang, J, Zhang, S., and Yu, C. 2013. The spurious regression of AR (p) infinite-variance sequence in the presence of structural breaks. Computational Statistics and Data Analysis, 67(2): 25-40. https://doi.org/10.1016/j.csda.2013.04.011

[16] Jönsson, K. 2008. The accuracy of normal approximation in a heterogeneous panel data unit root test. Statistical Papers, 49(3): 565-579. https://doi.org/10.1007/s00362-006-0033-4

[17] Khan, G. Y., Bashir, M., & Mehboob, S. (2019). Structural breaks and unit roots in selected macroeconomic series: Evidence from Pakistan. Paradigms, 13(2), 65-69.

[18] Kitov, I.O., and Kitov, O.I. 2008. Long-term linear trends in consumer price indices. Journal of Applied Economic Sciences, Volume III, Summer, 2(4): 101-112.

[19] Kruse, R. 2011. A new unit root test against ESTAR based on a class of modified statistics. Statistical Papers, 52(1): 71-85. https://doi.org/10.1007/s00362-009-0204-1.

[20] Mishra, S.K. 2008. A new method of robust linear regression analysis: Some Monte Carlo experiments. Journal of Applied Economic Sciences, Volume III, Fall, 3(5): 261-269.

[21] Leong, C.K. 2015. Response to the comment on testing for spurious and co integrated regressions: A wavelet approach. Journal of Applied Statistics, 42(8): 1770-1772. https://doi.org/10.1080/02664763.2015.1020006

[22] Nelson, C.R., and Kang, H. 1984. Pitfalls in the use of time as an explanatory variable in regression. Journal of Business and Economic Statistics, American Statistical Association, 2(1): 73-82. https://EconPapers.repec.org/RePEc:bes:jnlbes:v:2:y:1984:i:1:p:73-82

[23] Otero, J., and Smith, J. 2012. Response surface models for the Leybourne unit root tests and lag order dependence. Computational Statistics, 27(3): 473-486. https://doi.org/10.1007/s00180-011-0268-y

[24] Perron, P., and Rodríguez, G. 2016. Residuals‐based tests for co integration with GLS Detrended Data. The Econometrics Journal, 19: 84-111. https://doi.org/10.1111/ectj.12056

[25] Pesaran, M.H., Smith, L.V., and Yamagata, T. 2013. Panel unit root tests in the presence of a multifactor error structure. Journal of Econometrics, 175 (2): 94-115. https://doi.org/10.1016/j.jeconom.2013.02.001

[26] Phillips, P.C.B. 1987. Time series regression with a unit root. Econometrica, 55(2): 277−301.

[27] Phillips, P.C. 2014. Optimal estimation of co integrated systems with irrelevant instruments. Journal of Econometrics, 178(P2): 210-224. https://doi.org/10.1016/j.jeconom.2013.08.022

[28] Rehman, A.U., and Malik, I. 2014. The modified R a robust measure of association for time series. Electronic Journal of Applied Statistical Analysis, 7(1): 1-13. https://doi.org/10.1285/i20705948v7n1p1

[29] Ventosa-Santaulària, D., and Noriega, A.E. 2015. A simple solution for spurious regressions. Communications in Statistics-Theory and Methods, 45(19): 5561-5583. DOI: https://doi.org/10.1080/03610926.2014.948196

[30] Vinod, H. 2016. New bootstrap inference for spurious regression problems. Journal of Applied Statistics, 43(2):317-335. https://doi.org/10.1080/02664763.2015.1049939

[31] Wang, G., and Han, N-W. 2015. Spurious regressions in time series with long memory'. Communications in Statistics-Theory and Methods, 44(4): 837-854. https://doi.org/10.1080/03610926.2012.753088

[32] Wang, J., and Jasra, A. 2016. Monte Carlo algorithms for computing\ alpha-permanents. Statistics and Computing, 26(1-2): 231-248. https://doi.org/10.1007/s11222-014-9491-z

[33] Yule, G.U. 1926. Why do we sometimes get nonsense-correlations between Time-Series? A study in sampling and the nature of time-series. Journal of the Royal Statistical Society, 89(1): 1-63. https://doi.org/10.2307/2341482