Onion Price Prediction Assisted by Exploratory Data Analysis Employing ARIMA Technique for Varanasi

Agricultural commodities, ranging from staple crops to specialized products, are crucial for food security, industrial raw materials, and rural economies. Predictive analysis, leveraging historical data and machine learning models, enables stakeholders to forecast trends, optimize agricultural processes, and anticipate market dynamics. Together, predictive analysis and visualization redefine agriculture by enhancing productivity, mitigating risks, and promoting sustainability. Work explores predictive analysis methodologies, techniques, and data sources in the context of agricultural commodities. Real-world examples highlight how visualizations assist decision-makers in the agriculture sector. Data employed for analysis comprises of 10,228 rows with data like year of production, place, price, and quantity for onion from cities across India, analysis was particularly done for Varanasi. An autoregressive integrated moving average (ARIMA) is employed which works statistically and uses time series data to predict future trends. Work is an effort to predict onion prices which have had a lot of variances lately taking into consideration dependent parameters, the predicted price has been evaluated with p value of 1.39e-10 and a critical value of e -3.47 for 1% which is at par with existing algorithms.