Forecasting area and production of black gram pulse in Bangladesh using arima models

This study attempted to shed light on the issues such as forecasting of area and production of black gram pulse in bangladesh. data on area and production of black gram pulse collected over a period of 47 years (1967-68 to 2013-14) from bangladesh bureau of statistics. autocorrelation function,partial autocorrelation function and phillips-perron unit root test are calculated for obtaining stationarity and identifying tentative autoregressive & moving average orders for the data. appropriate box-jenkins autoregressive integrated moving average model is fitted. the test statistic ljung-box q is used to check the model adequacy,jarque-bera test,is used to check the normality of the models while akaike information criterion and bayesian information criterion are used to determine the best forecasting model. arima (0,1,0) model was found suitable for both black gram pulse area and production. the performances of model were validated by comparing with percentage deviation from the actual values and mean absolute percent error (mape). finally,a forecast for the period 2014-15 to 2018-19 was made. © 2016,university of agriculture,all rights reserved.