Modeling and Forecasting Rice Production in Bangladesh: An Econometric Analysis

M. M. Miah

Abstract

The aim of this research is to find time series model and to forecast the rice production of Bangladesh. That is achieved by finding the tentative Autoregressive Integrated Moving Average (ARIMA) models that fit and forecast well for rice production in Bangladesh. There are mainly three types of rice that is cultivated in Bangladesh throughout the year as Aus, Aman and Boro. Season covering the whole country we have included all these three types of rice production for analysis. In this paper, we study the performance of ARIMA model. We apply the time series Autoregressive Integrated Moving Average (ARIMA) model of different lag order to model rice production in Bangladesh. The suitable and efficient model to represent the data of the time series is chosen according to the smallest values of AIC, BIC, RMSE and MAPE criteria are used to select the best ARIMA(p, d, q) model. Result shows that, the best selected ARIMA model for Aus, Aman and Boro productions are ARIMA (2, 1, 5), ARIMA (2, 1, 5) and ARIMA (1, 1, 1) respectively than other tentative ARIMA(p, d, q) models. We forecasted next seven years rice production and make a comparison between the original series and forecasted series which also shows the same manner indicating fitted model are statistically well behaved to forecast rice productions in Bangladesh. It is found from the analysis that ARIMA model gives better forecast more accurately in the short run.Keywords: ARIMA, Rice Production, BangladeshCite this Article M.M. Miah. Modeling and Forecasting Rice Production in Bangladesh: An Econometric Analysis. Research & Reviews: Journal of Statistics. 2019; 8(2): 10–28p.