Effect of rainfall in predicting Tomato prices in India: An application of SARIMAX and NARX model

Although weather variable is irrepressible source for predicting prices, but unfortunately it directly influences-the prices in market especially perishable commodities as it has created the demand-supply gap. The present study is an attempt to predicting tomato prices by using seasonal autoregressive moving average with exogenous variable (SARIMAX) and non linear autoregressive exogenous (NARX) model. These models are able to take advantage not only of historical data of tomato prices, but also of the impact of rainfall. It is observed from the results that the NARX model outperformed the SARIMAX model. The forecasting performance has been compared with respect to root mean square error (RMSE) and mean absolute percentage error (MAPE). The-study is an effort to predict the tomato prices by taking into account the important weather variable i.e., rainfall-so that the stakeholders may make production, marketing and policy decisions well in advance.