and dependent variables are in a linear relationship. While the
MRA model is simple and efficient, its prediction accuracy is low
[16e18]. Mohamed and Bodger [19] and Baker and Rylatt [20]
attempted to predict electricity consumption in New Zealand and
the UK domestic energy demand, respectively, using the MRA
model. The ARIMA model, a methodology developed by Box and
Jenkins, can predict future values using the auto-correlation analysis
of time series data [21]. If there is a seasonal pattern in time
series data, the SARIMA model can be used [22]. Gonzales et al. [23]
predicted the future values of five energetic variables (i.e., domestic
electricity consumption, industrial electric energy consumption black-coal production, anthracite production, and electric energy
production) in Northern Spain. Compared to exponential smoothing
and the MRA model, it was determined that the ARIMA model
was the most superior
and dependent variables are in a linear relationship. While theMRA model is simple and efficient, its prediction accuracy is low[16e18]. Mohamed and Bodger [19] and Baker and Rylatt [20]attempted to predict electricity consumption in New Zealand andthe UK domestic energy demand, respectively, using the MRAmodel. The ARIMA model, a methodology developed by Box andJenkins, can predict future values using the auto-correlation analysisof time series data [21]. If there is a seasonal pattern in timeseries data, the SARIMA model can be used [22]. Gonzales et al. [23]predicted the future values of five energetic variables (i.e., domesticelectricity consumption, industrial electric energy consumption black-coal production, anthracite production, and electric energyproduction) in Northern Spain. Compared to exponential smoothingand the MRA model, it was determined that the ARIMA modelwas the most superior
การแปล กรุณารอสักครู่..
