Random error (also called accidental error) refers to the difference between the measured results and the average results of a large number of repeated measurements to be measured. "The average result of a large number of repeated measurements on the same object to be measured" refers to the expected value to be measured or the average value of all possible measured values obtained under repeated conditions. In regression analysis, the difference between the measured value and the predicted value according to the regression equation is expressed as δ. Residual δ obeys normal distribution N(0, σ2). The ratio of δ to σ is called normalized residual, which is expressed by δ *. δ * follows the standard normal distribution n (0, 1). The probability that the standardized residuals of experimental points fall outside the (-2,2) interval is ≤0.05. If the standardized residual of an experimental point falls outside the (-2,2) interval, it can be judged as an abnormal experimental point with 95% confidence and will not participate in the regression line fitting.

The so-called residual refers to the difference between the actual observed value and the regression estimated value.