Forecasting cereal crops production using time series analysis in Ethiopia

Abstract

Cereal crops dominate crop production and human diet in Ethiopia. In the present study major cereal crops production data for Ethiopia was analyzed using time-series analysis. Major cereal crops production of years 2000/01 to 2020/21 at national level of private peasant holdings during “Meher” season in Ethiopia was used as input to forecast the cereal crops production up to the year 2020 to 2030. The non-parametric Mann Kendall trend test was used with PAST (Paleontological Statistics Software Package for Education and Data Analysis) software to conduct the trend test. The Box-Jenkins Aautoregressive Integrated Moving Average (or ARIMA) method was put into use to forecast the production. The result shows that grain production had a significant increasing trend. Despite the differences in magnitude, all crops showed an increasing trend in production over the years. Mann-Kendall trend test showed significant trend for all crops. ARIMA and any of the smoothing models were found good in terms of R2 and AIC. Predictions of cereal crops production using ARIMA between 2020 and 2030 years showed increasing trends in all crops. Teff was forecasted from 5.8 tons at the beginning (2020/1) to 7.8 tons at the end of the forecast period (2030). Similarly, maize was forecasted from about 11 to 14, wheat from 6 to 7.5, sorghum from 5 to 7, and barley from 2.3 to 2.7. Increments ranged in between about 20% to 40%.

Keywords

• Cereals
• Auto Correlation Function
• Patrial autocorrelation function
• ARIMA
• Forecasting
• JMP-software