The work aims to establish the calibration regression equation model for acetone in the packaging composite film, so as to improve the accuracy of the detection results of acetone. With respect to the residual solvent of acetone, based on the Method for the Determination of Residual Solvent in Packaging Materials YBB00312004—2015, the calibration regression equation model for acetone in the packaging composite film was established in the weighted least square method. Moreover, through the outlier test, correlation analysis, parameter estimation, and judging indexes, such as homoscedasticity, the calibration regression equation model was comprehensively evaluated and analyzed by comparing with the ordinary least square calibration regression equation. The weighted least square method could effectively eliminate the influence of the heteroscedasticity on the calibration regression equation, and the weighted least square calibration regression equation was of significant linearity and at high regression level, and its detection results were of high accuracy. In conclusion, the weighted least square calibration regression equation model is suitable for the detection of acetone in the packaging composite film.