Machine learning techniques for predicting the price of brinjal in markets of Odisha

Price forecasting of vegetables has important implications for farmers, traders as well as consumers. A timely and accurate forecast of price helps the farmers switch between alternative nearby markets to sell their products and get good prices. The farmers can use the information to make choices around the timing of marketing. Brinjal is one of the important vegetables consumed all over the country. For forecasting the price of agricultural commodities, several statistical models have been applied in past but those models have their limitations in terms of assumptions. Recently, Machine Learning (ML) techniques have been quite successful in modelling time series data. Though numerous empirical studies have shown that ML approaches outperform time series models in forecasting different financial assets, their application in forecasting vegetable prices in India is scarce. In the present investigation, an attempt has been made to apply efficient ML algorithms e.g. Generalized Neural Network (GRNN), Support Vector Regression (SVR), Random Forest (RF) and Gradient Boosting Machine (GBM) for forecasting the wholesale price of Brinjal in major markets of Odisha. An empirical comparison of the predictive accuracies of different models with that of the usual autoregressive integrated moving average (ARIMA) model is carried out and it is observed that ML techniques particularly GRNN perform better.