6533b85dfe1ef96bd12bdcdf
RESEARCH PRODUCT
Pointwise resolutive significance of data and applications in experimental design and data treatment
Guillermo Ramis-ramosCarlos Mongay-fernándezJuan José Baeza-baezasubject
PointwiseData processingChemistryRegression analysisBiochemistryMeasure (mathematics)Standard deviationAnalytical ChemistryData setStatisticsEnvironmental ChemistryPoint (geometry)AlgorithmSpectroscopyReciprocaldescription
Abstract The concept of the resolutive significance of a point in a data set with regard to a number of addressed parameters is introduced, and two algorithms able to measure it are proposed. The algorithms are validated using simulated experiments. The sum of all the pointwise resolutive significances of a data set is also proposed as a measure of the resolution of the data set. This sum correlates well with the reciprocal of the standard deviation of the fitted parameters, indicating the precision that can be expected for each parameters. Applications in experimental design, and a method for establishing the weights in the least-quarters regression analysis are discussed.
| year | journal | country | edition | language |
|---|---|---|---|---|
| 1992-08-01 | Analytica Chimica Acta |