Time series analysis of maize prices in Kampala region Uganda (from 2010 to 2019)

Abstract

The major objective of the study was to carry out a time series analysis of maize prices in Kampala Uganda for the period 2010 to 2019 and the predict prices for the next five years. The data was obtained from the Uganda bureau of statistics for months of July 2010 to April 2019 for Kampala region In this study, we investigated the change in the monthly and yearly trends of maize flour prices in Kampala from 2010 to 2019 to determine if there is any increase or decrease thus fluctuations in prices. A trend analysis was applied to the monthly maize prices fitted to their appropriate ARIMA model. The trend still showed that there are unstable prices as years and months go by many differences in weather and bumper harvests In accordance with ARIMA model assumptions of time series: white noise assumes that the residuals have zero mean, constant variance and the autocorrelation of any observations of such sequence is always zero (uncorrelated). With the help of the ACF and PACF plots, tentative models were fit to the data. ARIMA (1, 2, 0) was noted to fit the data well. Further adequacy test on the model also confirmed the validity of the selected model. The model was then used to forecast for monthly prices of maize for the next period where the series continues to show an upward trend meaning that prices are rising. In conclusion, the provided forecasts will enable farmers who have been incurring huge losses due to uncertain prices to take appropriate acreage, selling and storage decisions. For example, they should be ready for the rising prices in the future and so they take on proper storage decisions The findings of this research recommend the Government of Uganda (GOU) and non-governmental organizations to understand patterns of maize prices and implement price stabilization strategies such as supporting the construction of silos, warehouses, and establishing commodity funds.