Investigating Instability of Regression Parameters and Structural Breaks in Nigerian Economic Data from 1984 to 2019

Nwakuya Tobechukwu Maureen; Biu Emmanuel Oyinebifun; Ekwe Christopher

doi:https://doi.org/10.14445/22315373/IJMTT-V68I12P509

International Journal of Mathematics Trends and Technology

Research Article | Open Access | Download PDF

Volume 68 | Issue 12 | Year 2022 | Article Id. IJMTT-V68I12P509 | DOI : https://doi.org/10.14445/22315373/IJMTT-V68I12P509

Investigating Instability of Regression Parameters and Structural Breaks in Nigerian Economic Data from 1984 to 2019

Nwakuya Tobechukwu Maureen, Biu Emmanuel Oyinebifun, Ekwe Christopher

Received	Revised	Accepted	Published
26 Oct 2022	01 Dec 2022	14 Dec 2022	31 Dec 2022

Citation :

Nwakuya Tobechukwu Maureen, Biu Emmanuel Oyinebifun, Ekwe Christopher, "Investigating Instability of Regression Parameters and Structural Breaks in Nigerian Economic Data from 1984 to 2019," International Journal of Mathematics Trends and Technology (IJMTT), vol. 68, no. 12, pp. 67-73, 2022. Crossref, https://doi.org/10.14445/22315373/IJMTT-V68I12P509

Abstract

Instability of regression coefficients can be an indication of structural break. This research investigated the existence of structural breaks and regression parameter instability using the Quandt Likelihood Ratio Test and the CUSUM Test. The Nigerian Real Gross Domestic Product was regressed on health and agricultural expenditure from 1984 to 2019. Five regression models were employed in this study (Linear, Logarithm, Inverse, Power and Exponential). The results revealed that the Quandt Likelihood Ratio (QLR) and CUSUM test identified instability of the regression parameters and structural break at different points for each of the models considered while the Harvey-Collier test was able to show nonlinearity in all the models, hence confirming existence of structural breaks. It was observed that both tests were seen to identify close structural breaks using the inverse model. The cumulated sum of scaled residuals shows that the inverse model and the exponential model’s scaled residual exhibited similar distribution patterns while linear, logarithm and power multiple regression models showed the same scaled residuals distribution behavior. Some of the structural breaks were seen 1998 which was the year the military regime ended, 1988, 2007, 2010 and 2015 which were the periods of change of various civilian regimes. In conclusion the study was able to show that the economic data considered requires to be spilt in other to avoid erroneous prediction of Nigeria Gross Domestic Product (GDP) based on agricultural and health expenditures.

Keywords

Cusum test, Harvey-Collier test, Quandt likelihood ratio Test, Regression parameter instability and Structural break.

References

[1] Donald W. K. Andrews, “Tests for Parameter Instability and Structural Change with Unknown Change Point,” Econometrica, vol. 61, no. 4, pp. 821-856, 1993. Crossref, https://doi.org/10.2307/2951764
[2] Donald W. K. Andrews, and Werner Ploberger, “Optimal Tests when a Nuisance Parameter is Present Only under the Alternative,” Econometrica, vol. 62, no. 6, pp. 1383-1414, 1994. Crossref, https://doi.org/10.2307/2951753
[3] Jushan Bai, Robin L. Lumsdaine, and James H. Stock, “Testing for and Dating Breaks in Multivariate Time Series,” Review of Economic Studies, vol. 65, pp. 395-432, 1998.
[4] O. Emmanuel Biu, and T. Maureen Nwakuya, “Chow Test for Structural Break: A Consideration of Government Transition in Nigeria form Military to Civilian Democratic Government,” Probability Statistics and Econometric Journal, vol. 4, no. 1, pp. 14-19, 2022.
[5] R. L. Brown, J. Durbin, and J. M. Evans, “Techniques for Testing the Constancy of Regression Relationships over Time,” Journal of the Royal Statistical Society. Series B (Methodological), vol. 37, no. 2, pp. 149-163, 1975. Crossref, https://doi.org/10.1111/j.2517-6161.1975.tb01532.x
[6] Central Bank of Nigeria, “Statistical Records on Gross Domestic Product (GDP) Constant Basic Prices (Naira Million) Quarterly Data 1960 to 2019,” Statistical Bulletin, vol. 30, no. 19, pp. 1-25, 2019.
[7] Gregory C. Chow, “Tests of Equality between Sets of Coefficients in Two Linear Regressions,” Econometrica, vol. 28, no. 3, pp. 591-605, 1960. Crossref, https://doi.org/10.2307/1910133
[8] Vicente Esteve, Manuel Navarro-Ibáñez, and María A.Prats, “The Spanish Term Structure of Interest Rates Revisited: Cointegration with Multiple Structural Breaks 1974–2010,” International Review of Economics & Finance, Elsevier, vol. 25, pp. 24-34, 2013. Crossref, https://doi.org/10.1016/j.iref.2012.04.007
[9] Bruce E. Hansen, “Testing for Structural Change in Conditional Models,” Journal of Econometrics, vol. 97, no. 1, pp. 93-115, 2000. Crossref, https://doi.org/10.1016/S0304-4076(99)00068-8
[10] David V. Hinkley, “Inference about the Change-Point in a Sequence of Random Variables,” Biometrika, vol. 57, no. 1, pp. 1–17, 1970. Crossref, https://doi.org/10.1093/biomet/57.1.1
[11] Mohitosh Kejriwal, and Pierre Perron, “The Limit Distribution of the Estimates in Co-Integrated Regression Models with Multiple Structural Changes,” Journal of Econometrics, vol. 146, no. 1, pp. 59-73, 2008. Crossref, https://doi.org/10.1016/j.jeconom.2008.07.001
[12] E. S. Page, “Continuous Inspection Schemes,” Biometrika, vol. 41, no. 1, pp. 100–115, 1954. Crossref, https://doi.org/10.2307/2333009
[13] Ruibing Qin et al., “Strong Convergence Rate of Robust Estimator of Change Point,” Mathematics and Computers in Simulation, vol. 80, no. 10, pp. 2026-2032, 2010. Crossref, https://doi.org/10.1016/j.matcom.2010.02.012
[14] Richard E. Quandt, “The Estimation of the Parameters of a Linear Regression System Obeying Two Separate Regimes,” Journal of American Statistics Association, vol. 53, no. 284, pp. 873-880, 1958. Crossref, https://doi.org/10.2307/2281957
[15] Z. Jhou, and S. Y. Liu, “Inference for Mean Change-Point in Infinite Variance AR(p) Process,” Statistics & Probability Letters, vol. 79, no. 1, pp. 6-15, 2009. Crossref, https://doi.org/10.1016/j.spl.2008.05.040
[16] Iliyan Georgiev et al., "Testing for Parameter Instability in Predictive Regression Models," Journal of Econometrics, Elsevier, vol. 204, no. 1, pp. 101-118, 2018. Crossref, https://doi.org/10.1016/j.jeconom.2018.01.005
[17] Jukka Nyblom, “Testing for the Constancy of Parameters Over Time,” Journal of American. Statistics Association, vol. 84, no. 405, pp. 223-230, 1989. Crossref, https://doi.org/10.2307/2289867
[18] Bradley S Paye, and Allan Timmermann, “Instability of Return Prediction Models,” Journal of Empirical Finance, vol. 13, no. 3, pp. 274-315, 2006. Crossref, https://doi.org/10.1016/j.jempfin.2005.11.001
[19] Jushan Bai, and Pierre Perron, “Estimating and Testing Linear Models with Multiple Structural Changes,” Econometrica, vol. 66, no. 1, pp. 47–78, 1998. Crossref, https://doi.org/10.2307/2998540
[20] Paul Michels, and Götz Trenkler, “Testing the Stability of Regression Coefficients Using Generalised Recursive Residuals,” Australian & New Zealand Journal of Statistics, vol. 32, no. 3, pp. 293–312, 1990. Crossref, https://doi.org/10.1111/j.1467-842X.1990.tb01025.x
[21] S. Mustafa, K. Riaz, and Q. Perveen, “Stability of Linear Regression Models,” Science International, vol. 27, no. 1, pp. 73-76, 2014.
[22] René Garcia and Pierre Perron, “An Analysis of the Real Interest Rate Under Regime Shifts,” Review of Economics and Statistics, vol. 78, no. 1, pp. 111– 125, 1996. Crossref, https://doi.org/10.2307/2109851
[23] Jian Liu, Shiying Wu, and James V. Zidek, “On Segmented Multivariate Regressions,” Statistica Sinica, vol. 7, pp. 497– 525, 1997.
[24] Robin L. Lumsdaine, and David H. Papell, “Multiple Trend Breaks and the Unit Root Hypothesis,” Review of Economics and Statistics, vol. 79, no. 2, pp. 212– 218, 1997.
[25] K. Morimune, and M. Nakagawa, “Unit Root Tests Which Allow for Multiple Breaks,” Kyoto Institute of Economic Research, Kyoto University, 1997.