The prediction of the mechanical properties of hot rolled steel sheet is one of the most challenging tasks for the steel manufacturer. Typically, the chemical composition of raw material for hot rolling process is not stable. Therefore steel manufacturer has to predict the mechanical properties including yield strength (YS), tensile strength (TS) and elongation (EL) based on input chemical composition. Currently, the method for prediction is the simple linear regression (SLR). The predictor in SLR is only the carbon equivalent which is inadequate to accurately predict the mechanical properties. Thus, this paper proposes the multiple linear regression (MLR) to determine more accurate regression equation. There are three important rolling process parameters considered in MLR including sheet thickness, finish temperature, and coiling temperature. It is found that MLR can yield better prediction value in terms of yield strength and elongation while SLR is still the appropriate method for the prediction of tensile strength. Based on the comparisons, both SLR and MLR can enhance the accuracy in predicting the mechanical properties of hot rolled steel sheet.