Price forecasting & anomaly detection for agricultural commodities in India

Fluctuations in food prices can cause distress among both consumers and producers, and are often exacerbated by trading networks especially in developing economies where marketplaces may not be operating under conditions of perfect competition for various contextual reasons. We look at onion and potato trading in India and present the evaluation of a price forecasting model, and an anomaly detection and classification system to identify incidents of hoarding of stock by the traders. Our dataset is composed of time series of wholesale prices and arrival volumes of the agricultural commodities at several village-level marketplaces, and retail prices of the commodities at the city centers. We also provide an in-depth qualitative analysis of the effect on these time series of events such as hoarding, weather disturbances, and external shocks. Our results are encouraging and point towards the possibility of building pricing models for agricultural commodities which can be used to reduce information asymmetries and to detect anomalies that can help regulate agricultural markets to operate more fairly.