An overview on multiple linear regression (MLR) is envisaged in this paper. All but the final section is devoted to a discussion of the basic concepts of MLR. The corresponding MLR equations are derived and presented in a useful form for computing. However, the entirely general matrix approach to least squares applicable to any linear regression situation is also envisaged. In the final section selected analytical and physicochemical applications are shown in tabular form. MLR is one of the most widely used statistical tool and found applications on a number of areas such as quantitative structure property relationships (QSPR), quantitative structure retention relationships (QSRR), quantitative structure-transformation relationships (QSTR), molecular linear free energy relationships (MLFER) and quantitative structure activity relationships (QSAR), solvent polarity and solvatochromic effects, parameter estimation methods, correction of spectral (matrix) interferences, prediction, modelling and optimization, Fourier transform near infrared spectroscopy (NIR-FT) and multicomponent spectrophotometric analysis, and in many other areas.