|
Abstract:
|
Chemical measurements obtained using liquid chromatography or other similar techniques are analyzed using a set of so-called measurement equations which relate the value being measured to various other variables, either measured, or partially known, based on properties of the measurement system such as calibration standards. In addition, the measurements are affected by various factors due to the experimental design. Thus the statistical analysis is necessarily quite complex. This presentation describes an analysis approach which incorporates the equations as well as the experimental factors in a Bayesian hierarchical model. The method is illustrated using measurements of 25-hydroxyvitamin D3 in solution reference materials via liquid chromatography with UV absorbance detection (LC-UV) and liquid chromatography mass spectrometric detection using isotope dilution (LC-IDMS).
|
