A novel generalized least square (GLS) estimator program was employed for determination of styrene (STY)/butyl acrylate (BA) reactivity ratios synthesized by solution copolymerization method. The monomer reactivity ratios as well as the 95 % individual confidence limits were determined by application of conventional linear methods like Finemann-Ross, Ezrielev-Brokhina-Roskin, Joshi-Joshi, Kelen-Tudos, modified Kelen-Tudos, extended Kelen-Tudos and Mao-Huglin. The estimation process was performed by applying techniques based on ordinary least square (OLS) and GLS approaches and the results were compared. The results showed a fairly good agreement between the experimental and calculated copolymer compositions. The model was then successfully validated through handling regression models with error terms that are heteroskedastic or autocorrelation, or both and clearly showed that the model was able to predict the reactivity ratios by accounting response error structure. Based on the copolymer compositions determined by 1H NMR, the reactivity ratios of STY and BA were found to be 0.886634 and 0.216369, respectively, by Mao-Huglin method through the GLS approach, and this new estimation method shows the best linear estimations for the monomer reactivity ratios. The present paper shows a new estimation integral approach for determining the monomer reactivity ratios by different conventional linear methods at low and high conversions in E Views software and the calculated values are discussed in terms of regression models.