Least-squares fitting of parameters
In crystallography, numerical parameters for the structure are derived from experimental data. This chapter discusses how the data and parameters are related, and introduces data fitting procedures including unweighted and weighted means, and least-squares criteria for a ‘best fit’. The simple case of linear regression for two parameters of a straight line is treated in some detail in order to explain the least-squares tools of observational equations and matrix algebra, leading to variances and covariances. Restraints and constraints are applied, and their important distinction made clear. Non-linearity in the observational equations leads to further complications, with only parameter shifts rather than the parameters themselves obtainable through least-squares treatment. Ill-conditioning and matrix singularity are explained, with reference to crystallographic relevance. Computing aspects are considered, since least-squares refinement is particularly expensive computationally.
Keywords: least-squares refinement, data, parameters, weights, variance and covariance, linear least-squares, non-linear least-squares, restraints, constraints, ill-conditioning, singularity
Oxford Scholarship Online requires a subscription or purchase to access the full text of books within the service. Public users can however freely search the site and view the abstracts and keywords for each book and chapter.
Please, subscribe or login to access full text content.
If you think you should have access to this title, please contact your librarian.
To troubleshoot, please check our FAQs , and if you can't find the answer there, please contact us .