Application of Interrupted Time Series Modelling to Prime Motor Spirit Distribution in Nigeria

Ette Harrison Etuk, Imo Udo Moffat, Azubuike Samuel Agbam

Abstract


An inspection of the time-plot of monthly Prime Motor Spirit (PMS) distribution in Nigeria from 2009 to 2015 reveals an abrupt jump in January 2013 with the series continuing at that level till 2015. Clearly the trend of the series was interrupted in January 2013 and it is believed that this perturbation was due to the deregulation of the downstream sector of the crude oil industry.  A t-test comparison of the pre- and the post-intervention means is highly significant (p < 0.0001) indicating the impact of the intervention. A model of the ARIMA family was to be fitted to the pre-intervention data which were observed to have a downward trend and be non-stationary. Differencing once rendered it stationary. An adequate ARIMA(2,1,0) model was fitted to the original pre-intervention series. Post-intervention forecasts were obtained on the basis of this model. These forecasts were subtracted from their respective post-intervention counterpart observations. These differences were modelled to obtain the transfer function of the intervention. The resultant intervention model closely fits the post-intervention data and may be used to explain and control the situation.

Full Text:

PDF

References


G. E. P. Box and G. C. Tiao, “Intervention Analysis with applications to economic and environmental problems”, Journal of American Statistical Association, vol. 70, pp. 70-79, 1975.

J. B. Guerard, “Automatic time series modeling, intervention analysis, and effective forecasting,”Journal of Statistical Computation and Simulation., vol. 34, no. 1, pp. 43-49, 1988.

U. Helfenstein, “The use of transfer function models, intervention analysis and related time series methods in epidemiology,” International Journal of Epidemiology., vol. 20, no. 3, pp. 808-815, 1991.

S. Sridharan, “Intervention time series analysis of crime rates: the impact of sentence reforms in Virginia,” https://papers.tinbergen.nl/03040.pdf accessed 26 July 2017.

M. Ray, V. Ramasubramanian, A. Kumar and A. Rai, “Application of time series intervention modeling for modeling and forecasting cotton yield,”Statistics and Applications, vol. 12, nos. 1 & 2, pp. 61-70, 2014.

B. U. Oreko, C. C. Nwobi-Okoye, S. Okiy and A. C. Igboanugo, “Modelling the impact of intervention measures on total accident cases in Nigeria using Box-Jenkins methodology: A case study of Federal Road Safety Commission,” Cogent Engineering, vol. 4, no. 1., 2017.

R. Bell, S. V. Gliniaia, Z. Waal, A. Close, E. Moloney, S. Jones et al. “Evaluation of a complex healthcare intervention to increase smoking cessation in pregnant women: interrupted time series analysis with economic evaluation, www.tobaccocontrol.bmj.com/content/tobaccocontrol/early/2017/02/10/tobaccocontrol-2016-053476.full.pdf accessed on 26 July 2017.

E. H. Etuk and U. A. Victor-Edema, “Box-Tiao intervention modeling of monthly EUR-NGN exchange rates due to Nigerian economic recession,” Journal of Science and Engineering Research, vol. 4, no. 3, pp. 59-65, 2017.

The Pennsylvania State University, Welcome to STAT 510! Applied Time Series Analysis. Department os Statistics Online Program. www.onlinecourses.science.psu.edu/stat510/ accessed 9th November 2016.


Refbacks

  • There are currently no refbacks.




© International Journals of Advanced Research in Computer Science and Software Engineering (IJARCSSE)| All Rights Reserved | Powered by Advance Academic Publisher.