Accelerating Linear Regression: A Performance Evaluation Using Sequential, Multicore, and Parallel Computing

Linear Regression is one of the most fundamental algorithms in the field of machine learning and statistical data analysis, offering a simple yet powerful approach for predictive modeling. This paper presents an extensive study of the Linear Regression algorithm, emphasizing its theoretical foundation, implementation process, and practical applications in prediction tasks. The research explores the mathematical formulation of the model, where the relationship between dependent and independent variables is expressed through a linear equation, and parameters are estimated using the Ordinary Least Squares (OLS) method to minimize prediction errors. Experimental analysis conducted on both synthetic and real-world datasets demonstrates that Linear Regression provides reliable and interpretable predictions when the underlying relationship between variables is linear. However, its accuracy declines for complex nonlinear data patterns. Despite these limitations, Linear Regression remains a cornerstone in predictive analytics due to its computational efficiency, interpretability, and ability to serve as a foundational framework for developing more advanced regression and machine learning algorithms