Comparative Study of District Level Rice Production under Artificial Neural Network and Multiple Linear Regression Model

Abstract

Using Artificial Neural Network (ANN) methodology a district level prediction of rice production in Uttarakhand state of India has been proposed. The proposed model has been empirically compared with the existing Multiple Linear Regression (MLR) Model. Analysis of data obtained from a survey carried out from Directorate of Economics and Statistics, Uttarakhand, India revealed that ANN methodology performs better as compared to MLR model in terms of R- square value, Mean Square Error (MSE) value and Root Mean Square Error (RMSE) values. This approach does not require any additional survey or conducting extra crop cutting experiments for crop production estimate at the district level.