# Difference between revisions of "Maximum likelihood"

### From Online Dictionary of Crystallography

BrianMcMahon (talk | contribs) |
m (lang) |
||

Line 1: | Line 1: | ||

+ | <font color="blue">Maximum de vraisemblance</font> (''Fr''). <font color="red">Maximale Wahrscheinlichkeit</font> (''Ge''). <font color="green">Máxima verosimilitud</font> (''Sp''). <Font color="black">Metodo della massima verosimiglianza</Font> (''It''). <Font color="purple">最尤法</Font> (''Ja'') | ||

+ | |||

+ | |||

An approach to structure [[refinement]] in which the parameters of a structural model are modified to optimize the statistical probability of generating a set of observed intensities. The technique is often used in the refinement of structures of biological macromolecules, where the unfavourable parameter-to-observation ratio often leads to overfitted data and consequent systematic errors in least-squares minimization procedures. | An approach to structure [[refinement]] in which the parameters of a structural model are modified to optimize the statistical probability of generating a set of observed intensities. The technique is often used in the refinement of structures of biological macromolecules, where the unfavourable parameter-to-observation ratio often leads to overfitted data and consequent systematic errors in least-squares minimization procedures. | ||

[[Category:Biological crystallography]] | [[Category:Biological crystallography]] | ||

[[Category:Structure determination]] | [[Category:Structure determination]] |

## Revision as of 14:14, 20 March 2015

Maximum de vraisemblance (*Fr*). Maximale Wahrscheinlichkeit (*Ge*). Máxima verosimilitud (*Sp*). Metodo della massima verosimiglianza (*It*). 最尤法 (*Ja*)

An approach to structure refinement in which the parameters of a structural model are modified to optimize the statistical probability of generating a set of observed intensities. The technique is often used in the refinement of structures of biological macromolecules, where the unfavourable parameter-to-observation ratio often leads to overfitted data and consequent systematic errors in least-squares minimization procedures.