Price forecasting models for rice in West Bengal, India

Rice “a staple crop in India” plays a pivotal role in ensuring food security and sustenance for a significant portion of the global population. This study focuses on the development and evaluation of price forecasting models for rice in West Bengal, India, a region renowned for substantial rice production. Accurate price forecasts are crucial for various stakeholders, including farmers, policymakers, and agribusiness industries. The research assesses a range of forecasting methods, including traditional statistical models: AutoRegressive Moving Average (ARMA) and AutoRegressive Integrated Moving Average (ARIMA) and modern machine learning models: Recurrent Neural Networks (RNN), Artificial Neural Networks (ANN), Gated Recurrent Unit (GRU), Long Short-Term Memory (LSTM), and Stacked Long Short-Term Memory (SLSTM). The evaluation metrics such as Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) were used to measure accuracy of the models. The scores of evaluation metrics were used to select the best model. The results show that machine learning models outperform traditional statistical models, with Stacked LSTM being the most suitable for accurately forecasting rice prices as it had lowest RMSE (8.718) and MAPE (0.010) values. These insights are valuable for decision-making within the rice market, aiding farmers in optimizing production and assisting traders and policymakers in making profitable decisions and formulating effective policies to ensure food security and economic stability.