Onion Price Forecasting using Machine Learning Models Market of South Punjab

Abstract

Price plays a crucial role in financial activities, with unexpected fluctuations often signaling market instability. In today’s market, machine learning offers a variety of techniques to forecast commodity prices, helping to manage such instability. This paper explores the application of machine learning methods for predicting onion prices, using data obtained from the Ministry of Agriculture, South Punjab, Pakistan. We applied machine learning algorithms such as Linear Regression, SARIMA, LSTM, SVR, and Random Forest Regression to make predictions. We then evaluated and compared the performance of these techniques to determine which provides the highest accuracy. Our findings indicate that all the methods used to determine which offer high accuracy, and suggest that all the techniques produced similar results. Using these methods, we aim to forecast onion prices into three categories: low (preferable), medium (economical), and high (expensive).