Using ARIMA Models to Predict The Economic Variables of Oranges in Egypt | ||
| Scientific Journal of Agricultural Sciences | ||
| Volume 5, Issue 4, December 2023, Pages 227-250 PDF (1.63 M) | ||
| Document Type: Original Article | ||
| DOI: 10.21608/sjas.2023.235280.1337 | ||
| Authors | ||
| Magda Medhat Hessien* 1; M.A.A Abdalla2; Asmaa M.E.B3; Mohamed S.A.E3 | ||
| 1Department of Agricultural Economics, Faculty of Agriculture, Banha University | ||
| 2Department of Agricultural Economics, Faculty of Agriculture, Ain Shams University | ||
| 3Department of Agricultural Economics, Faculty of Agriculture, Benha University. | ||
| Abstract | ||
| This study aims to use ARIMA (Autoregressive Integrated Moving Average) models to forecast economic variables for the orange crop in Egypt. Time series are considered essential statistical methods that allow understanding the trends and changes in economic values over time and assist in determining methods, results, and interpreting the observed relationships. They also enable forecasting future changes in values based on past data, which is beneficial in formulating future economic policies and plans for the country. Therefore, this research primarily targeted predicting economic variables for the orange crop in Egypt by identifying the most suitable standard methods for forecasting. The results revealed that the total area of orange cultivation in Egypt during the period from 2007 to 2021 fluctuated between two extremes, with the lowest being approximately 248.233 thousand acres in 2007 and the highest reaching around 378.107 thousand acres in 2015. The forecast results for the total area indicated an increasing trend over time, with occasional declines occurring at different time intervals. The results also suggest that the time series may not be stable on average. The use of Auto Correlation Function (ACF) and Partial Correlation was applied to detect the stability of the time series, indicating that it is nonstationary. Unit Root Tests (Dickey-Fuller) were employed to test the stability of the time series, and the results confirmed its stability, as the p-values for the t-test were less than 5% and 1%, indicating statistical significance and stability in the series, whether with or without the constant and trend. | ||
| Keywords | ||
| ARIMA models; Production; Moving Time Series; Orange | ||
|
Statistics Article View: 98 PDF Download: 163 |
||
