Forecasting Electricity Generation in Kurdistan Region Using BOX-Jenkins Model

Abstract

The objective of this research is to identify the best and most relevant statistical model for projecting electrical power generation in the KRI. Data was collected for this purpose throughout a 168-year period (2006-2019). The Box-Jenkins technique was used, and it was discovered that the series is unstable and not random after analyzing it. The essential transformations, namely the square root and the first difference, were used to achieve stability and randomization. The necessary transformations, such as the square root and the first difference, were used to achieve stability and randomness. the analysis showed that ARIMA (2,1,2) is the most appropriate model among the proposed models using some statistical criteria like (AIC, BIC, MSE, MAPE, and RMSE) were used to obtain the model that can be utilized in the prediction. A simulation was conducted in favor to the selected model.

Keywords

• electricity generation
• time series
• box-jenkins
• forecasting
• simulation