Error analysis of EXAFS measurements

Abstract

A method to analyze EXAFS data is proposed which follows the error propagation from the measured input data to the output of the physical parameters taking into account experimental and systematic errors. The Bayesian approach is used to describe the modification of the a priori expectation for the model parameters due to the measurement. A description is discussed to determine the relative weight of the a priori information compared to the experimental information in their impact on the results for the a posteriori expectation values and their variances. Computer generated data above the Ta L3-edge are used to demonstrate the robustness of the method.