APPROPRIATE CALIBRATION FUNCTIONS FOR CAPILLARY ELECTROPHORESIS .2. HETEROSCEDASTICITY AND ITS CONSEQUENCES

If ordinary least squares regression methods are to be used, the stand ard deviation of the signal should not depend on the sample concentrat ion, but this is not true in CE. Results indicate, that the signal sta ndard deviation is approximately proportional to the sample concentrat ion. Therefore weighted least squares regression must be used, if the standard deviation within the concentration range differs by more than the factor 50. It is advised to use this regression method down to th e factor 5, where the difference to ordinary least squares calculation s is still. significant. This is demonstrated by comparing experimenta l and simulated data. These considerations are valid for other analyti cal techniques as well, if their characteristics of calibration and va riance function are similar.