ARMA MODEL FOR PREDICTION OF SOLAR RADIATION IN KANO-NIGERIA

Authors

  • I. S. Madugu Department of Electrical Engineering; Kano University of Sci. & Tech.
  • B. J. Olufeagba Department of Electrical and Electronics Engineering, University of Ilorin
  • Y. A. Adediran Department of Electrical and Electronics Engineering, University of Ilorin
  • A. Abdulkarim Department of Electrical Engineering, Ahmadu Bello University Zaria
  • F. Abdulkadir Department of Chemistry Education, Kano University of Sci. and Tech.
  • J. U. Inaku Department of Electrical and Electronics Engineering, University of Ilorin
  • O. Ogunbiyi Department of Electrical and Computer Engineering, Kwara State University
  • O. Ibrahim Department of Electrical and Electronics Engineering, University of Ilorin
  • A. U. Lawan Department of Electrical Engineering, Federal Polytechnique Kazaure
  • M. A. Afolayan Department of Electrical and Electronics Engineering, University of Ilorin

Keywords:

Solar radiation, Time Series Box-Jenkins, Root Mean Square Error, and Sum of Square Error.

Abstract

Statistical approach is one of the most prominent ways of forecasting a trend of a time
series data such as solar radiation. This paper explore the autoregressive integrated
moving average (ARMA) model using Box and Jenkins methodology to determine the most
parsimonious model of Kano, Nigeria solar radiation time series. The result obtained
showed that ARMA(3,0) has the least value of Root Mean Square Error (RMSE), Sum Of
Square (SSE), Mean Absolute Percentage Error (MAPE) and Theil’s U-Statistics with
0.0004, 0.0422, 0.0388 and 0.0061 respectively. Therefore ARMA(3,0) is the best model
that fits the solar radiation data and hence the most parsimonious.

Published

2019-06-27

Issue

Vol. 8 No. 1 (2019)

Section

Articles