Sugarcane Yield Estimation of Yamunanagar and Panipat District of Haryana Using NARX Model

Abstract

Uncontrollable factors that affect yield variability include seasonal and intra seasonal weather variations. The degree to which meteorological factors affect agricultural productivity depends not only on their quantity but also on how they are distributed throughout the crop season. Therefore, while estimating the dynamic behavior of agricultural production, it should be able to take advantage from both historical data on crop yield and the impact of different external environmental driving forces. In this study, Yamunanagar and Panipat districts of Haryana's sugarcane yield have been attempted to be estimated using NARX (Nonlinear Autoregressive Model with Exogenous Input) models. The NARX model with one hidden neuron and arf13, and the NARX model with two hidden neuron and tmx13 selected as best model for Yamunanagar and Panipat districts respectively with smaller values of percent relative deviation (RD (%)), mean absolute percentage error (MAPE) and root mean squared error (RMSE).