Some applications of statistical tests in analysis of EXAFS and SEXAFS data

Abstract

A simple formula is developed, based on standard statistical methods, that can be used to examine the physical significance of adding a new shell of neighbors to an EXAFS analysis. This can readily be applied in conjunction with the widely used data analysis program EXCURVE. The formula is appropriate for fitting primary data only, but a variation is offered for use in fitting data that have been forward and reverse Fourier transformed. The problem of error estimation where correlated parameters are involved is also considered.