Employing the principal components in time series models and selecting the best models with Application

In this study, principal components analysis, which is one of the methods of multivariate analysis for prediction of time series models (Box-Jenkins Model) was used by applying to electric power data (Erbil Gas Power Plant) (EGPS) which contains multivariate data (5 stations) and the data was month...

Full description

Saved in:
Bibliographic Details
Main Authors: Rizgar Maghded Ahmed, Nida Salim Mala Younis
Format: Article
Language:Arabic
Published: Salahaddin University-Erbil 2023-04-01
Series:Zanco Journal of Humanity Sciences
Online Access:https://zancojournal.su.edu.krd/index.php/JAHS/article/view/552
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this study, principal components analysis, which is one of the methods of multivariate analysis for prediction of time series models (Box-Jenkins Model) was used by applying to electric power data (Erbil Gas Power Plant) (EGPS) which contains multivariate data (5 stations) and the data was monthly for the period from (1/1/2017) to (14/9/2021). The idea of ​​the research was based on applying principal component analysis to multiple time series data, obtaining the components extracted from them, and then estimating the Box-Jenkins Models. The main conclusion is that principal component analysis is effective in reducing multiple time series data and obtaining the best models based on statistical criteria. And finally, the best proposed model for predicting electrical energy production data in the City of Erbil is (ARIMA(2,2,2)x(2,2,0)12).
ISSN:2412-396X