PROBABILISTIC APPROACH TO CONFIDENCE-INTERVALS OF LINEAR CALIBRATION

It is demonstrated that the statistical reliability of experimentally determined confidence intervals in some linear calibration problems in instrumental analyses can greatly be enhanced if the standard deviati on (s, sigma(M)) of measurements, which is a theoretical prediction of the instrumental response error, is incorporated into the usual stati stical equation instead of the residual of least-squares fitting, The reliability of the calibration depends on not only the fluctuation of calibration lines due to the response error, sigma(M), but also the er ror of parametrization associated with the a priori response error pre diction, i.e., the variance of the response st Var(sigma(M)). The erro r prediction requires the Fourier transform of an instrumental baselin e, signal shape and others (e.g., sample injection error). The variabi lity in the confidence interval from five calibration standards is as small in the probabilistic approach as that for 50 standards in the st atistical method, Liquid chromatography and capillary electrophoresis are taken as examples.