Fitting the Data

Abstract

Fitting models to experimental data is done by nonlinear regression routines, which vary multiple parameters like rate constants, etc. until an optimal fit is achieved. The resulting set of parameters need not be unique. Parameters often are correlated, so that the variation of one parameter can be compensated by variations of other parameters without reducing the quality of the fit. A correlation matrix helps to clarify this point. Experimental procedures, strategies for the reduction of the number of varied parameters, and global fits enhance the reliability of derived rate constants. At the end of this chapter, the reader should be able to import data from a spreadsheet, fit them to any reaction scheme and do a critical assessment of the significance of the fitted values.