经典最小二乘法不考虑自变量的实验误差,而将因变量的残差平方和作为目标函数,由实验数据求出目标函数最小时的模型参数。然而选定的目标函数不同,所得的参数也有所改变。这说明这些参数具有不确定因素,不是最佳值。 最大似然原理是统计学的一种概念。它考虑到各种测量变量的随 Classical least square method does not consider the experimental error of independent variables, and takes the residual sum of squares of the dependent variables as the objective function, and obtains the model parameters when the objective function is the minimum from the experimental data. However, the selected objective function is different, the resulting parameters have changed. This shows that these parameters have uncertainties, not the best value. The principle of maximum likelihood is a concept of statistics. It takes into account the various measurement variables