Forecasting Agriculture Commodity Price Trend using Novel Competitive Ensemble Regression Model

This paper introduces a novel approach for forecasting the price trends of agricultural commodities to address the issue of price volatility faced by both farmers and consumers. The accurate forecasting of food prices is particularly crucial in emerging nations such as India where food security is a top priority. To achieve this goal, the paper presents an ensemble learning-based approach for predicting the agricultural commodity price (ACP) trend. Using dataset namely rainfall and wholesale pricing index (WPI), the study compares the performance of various individual and ensemble regression models. The findings of this work demonstrated that the novel competitive ensemble regression (CER) approach outperforms traditional individual regression models in predicting price fluctuations trend accurately. This approach has the high potential and more precise prediction to afford farmers and dealers, also make the model suitable for the financial industries.