Electroencephalography (EEG) provides a non-invasive method for analyzing brain activity and understanding cognitive and emotional responses to external stimuli such as music. This research investigates EEG-based brain responses to music using statistical analysis and machine learning techniques. The proposed methodology includes data preprocessing, descriptive statistical analysis, correlation analysis, hypothesis testing, and linear regression to examine relationships among EEG-derived features, including Enhance, Suppress, R-Enhance, and R-Suppress signals. Experimental results demonstrate statistically significant differences between enhancement and suppression signals while revealing the complex, non-linear characteristics of EEG data that limit simple linear predictive models. The findings contribute to neuroscience by improving the understanding of music-induced brain activity and provide valuable insights for applications in emotion recognition, brain-computer interfaces, cognitive neuroscience, and music therapy.