ASSESSING THE ACCURACY OF ANALYTICAL METHODS USING LINEAR-REGRESSION WITH ERRORS IN BOTH AXES

In this paper, a new technique for assessing the accuracy of analytica l methods using linear regression is reported, The results of newly de veloped analytical methods are regressed against the results obtained using reference methods, The new test is based on the joint confidence interval for the slope and the intercept of the regression line, whic h is calculated taking the uncertainties in both axes into account, Th e slope, intercept, and variances which are associated with the regres sion coefficients are calculated with bivariate least-squares regressi on (BLS), The new technique was validated using three simulated and fi ve real data sets, The Monte Carlo method was applied to obtain 100 00 0 data sets for each of the initial simulated data sets to show the co rrectness of the new technique, The application of the new technique t o five real data sets enables differences to be detected between the r esults of the joint confidence interval based on the BLS method and th e results of the commonly used tests based on ordinary least-squares o r weighted least-squares regression.