Farmers play an important role in agriculture. When crop prices drop after harvest, farmers face huge losses. Therefore, it is beneficial for farmers to know the predicted price of crops before planting. In this research, our aim is to study and compare the performance of multiple linear regression, decision tree regression, random forest regression, support vector regression, and deep learning regression algorithms for predicting cotton prices in the Rajkot district. We evaluated the performance of the models based on Mean Absolute Error (MAE), Mean Squared Error (MSE), and R-squared (R2). The results show that Linear Regression and Random Forest are the most accurate models, with Linear Regression offering the best performance overall. Random Forest also performed well, closely following Linear Regression. Deep Learning showed moderate performance, while Decision Tree also had good predictive ability, but with a slightly higher error rate. On the other hand, SVR showed a significantly higher error rate, making it unsuitable for the dataset. The correlation matrix reveals strong relationships among minimum, maximum, and modal prices. This analysis also indicates consistent and stable price behavior. This study offers a systematic analysis of regression models for cotton price prediction, aiming to provide guidance to farmers and stakeholders in making informed financial decisions for enhanced economic stability.