A SARIMA and Adjusted SARIMA Models in a Seasonal Nonstationary Time Series; Evidence of Enugu Monthly Rainfall
European Journal of Mathematics and Statistics · 2021
DOI: 10.24018/ejmath.2021.2.1.15 ↗Kurz gefasst
Fachartikel in der Zeitschrift European Journal of Mathematics and Statistics, 2021. Verfasst von Chibuzo Gabriel Amaefula. Die Arbeit wurde bisher 3-mal zitiert (Crossref). Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
The paper compares SARIMA and adjusted SARIMA(ASARIMA) in a regular stationary series where the underlying variable is seasonally nonstationary. Adopting empirical rainfall data and Box-Jenkins iterative algorithm that calculates least squares estimates, Out of 11 sub-classes of SARIMA and 7 sub-classes of ASARIMA models, AIC chose ASARIMA(2,1,1)12 over all sub-classes of SARIMA(p,0,q)x(P,1,Q)12 identified. Diagnostic test indicates absence of autocorrelation up to the 48th lag. The forecast values generated by the fitted model are closely related to the actual values. Hence, ASARIMA can be recommended for regular stationary time series with seasonal characteristics and where parameter redundancy and large sum of square errors are penalized.
Zitieren
APA
Amaefula, C. G. (2021). A SARIMA and Adjusted SARIMA Models in a Seasonal Nonstationary Time Series; Evidence of Enugu Monthly Rainfall. European Journal of Mathematics and Statistics. https://doi.org/10.24018/ejmath.2021.2.1.15
Vancouver
Amaefula CG. A SARIMA and Adjusted SARIMA Models in a Seasonal Nonstationary Time Series; Evidence of Enugu Monthly Rainfall. European Journal of Mathematics and Statistics. 2021. doi:10.24018/ejmath.2021.2.1.15
BibTeX
@article{Amaefula2021,
author = {Amaefula, Chibuzo Gabriel},
title = {A SARIMA and Adjusted SARIMA Models in a Seasonal Nonstationary Time Series; Evidence of Enugu Monthly Rainfall},
journal = {European Journal of Mathematics and Statistics},
year = {2021},
doi = {10.24018/ejmath.2021.2.1.15}
}
EUVIMED