With the current advancements in technology across various sectors and the increasing flow of data from different sources, the need for innovative methods in managing, analyzing, and forecasting business metrics has become crucial for companies seeking a competitive advantage to remain in the market while maximizing profits. This study explores the application of various machine learning and traditional statistical regression models to predict profit based on real-world datasets encompassing market dynamics for over 200 office products. Different factors, including price, cost of goods sold, advertising spend, stock levels, and competitor prices, have been considered to predict the profit of each product. Additionally, the study demonstrates the comparative effectiveness of using machine learning against traditional statistical regression models. The models considered in this study include Ordinary Least Squares and Generalized Least Squares as traditional regression models, and different machine learning models including Linear Regression, Lasso Regression, Ridge Regression, Random Forest, Gradient Boosting Regression, K-Nearest Neighbors Regression, and Decision Tree Regression. These models are evaluated based on performance metrics such as Mean Absolute Error, Mean Squared Error, Root Mean Squared Error, and R-squared. The evaluation results indicate that Random Forest performs the best, followed by Decision Tree Regression and then Gradient Boosting Regression. In the validation, the results of these measures show that Gradient Boosting Regression, followed by Random Forest, demonstrates the best prediction performance. Based on these findings, we can confirm that Gradient Boosting Regression and Random Forest models exhibit the best performance, offering superior prediction results. These results align with previous findings presented in the literature review, confirming that machine learning models have better prediction capability and accuracy than the traditional regression approaches.