This program offers the user a novel way to quantify the uncertainty of the calibration process. In general, calibration of an analytical procedure consists of three steps: (1) measuring the calibration points [x,y]_i and fitting the points by appropriate model function y=f(x), (2) measuring the signal, Yo, and (3) estimation of Xo=f ^-1(Yo) and its standard deviation SXo from the calibration curve. If the variance is assigned to the dependent variable, y, the classical least squares method is adequate for the estimation of Xo and SXo. However, this approach fails if the variance of the independent variable, y, needs to be considered. Calibration Pro allows the user to evaluate calibration curves when the variance of the independent variable (x) is significant. The method is based on a newly developed algorithm, the "rigorous least-squares parameter optimization algorithm", which provides estimates for Xo and its standard deviation by considering the standard deviations of the independent and the dependent variables. The model functions: line, parabola and cubic parabola are available for the least squares fitting. The Calibration Pro is equipped with a powerful management system designed to keep track all of calibration curves developed over time. The calibration curves recorded are protected against unauthorized use, but allows the database administrator to edit all records.

For more information on these products contact us at  info@chemsw.com

Return to Quality Control and Stats