ISSN:0886-9383    

DOI:10.1002/cem.1180030403

出版商:John Wiley&Sons, Ltd.    

年代:1989

数据来源: WILEY

摘要:

AbstractOne of the major application areas of factor analysis, multivariate calibration and quantitation, is covered in this review. The algorithms, methodologies and applications covered include principal component regression, target transformation factor analysis, singular value decomposition and rank annihilation factor analysis. Many important areas of research having relevance to multivariate calibration and quantitation problems are also covered in this review, including background correction, measurement error, rank determination, cross‐validation, figures of merit, detection of invalid samples, experimental design, sample selection, statistical inference and wavelength selectio

ISSN:0886-9383    

DOI:10.1002/cem.1180030404

出版商:John Wiley&Sons, Ltd.    

年代:1989

数据来源: WILEY

摘要:

AbstractData of the hydrolysis of adenosine 5′‐triphosphate have been re‐evaluated using a computer program that numerically integrates the differential rate equations within a routine that optimizes rate coefficients given a suitable model and concentration versus time data.The model is tested by calculation of HamiltonR‐values, the FisherF‐statistic, a sensitivity analysis, the standard errors on the rate coefficients and by constructing contour maps of the objective function versus two rate coefficients.An optimization using only phosphate concentration data cannot distinguish between a model in which adenosine 5′‐monophosphate is formed predominantly directly with a molecule of pyrophosphate, and one in which it is formed via adenosine 5′‐diphosphate. A more accurate set of rate coefficients is calculated from existing data and the relative importance of the two

ISSN:0886-9383    

DOI:10.1002/cem.1180030405

出版商:John Wiley&Sons, Ltd.    

年代:1989

数据来源: WILEY

摘要:

AbstractAnalysis of multivariate response data by modelling the principal components of the response has been applied to two sets of data. In both cases principal components analysis revealed the relationships among the response variables and exploited them to simplify the problem of modelling and optimizing the multivariate response. The models and optima obtained from the principal components compared favourably with the individual models and simultaneous optima.

ISSN:0886-9383    

DOI:10.1002/cem.1180030406

出版商:John Wiley&Sons, Ltd.    

年代:1989

数据来源: WILEY

摘要:

AbstractWhen the number of variables exceeds the number of samples, one method of multivariate discrimination is to use principal components analysis to reduce the dimensionality and then to perform canonical variates analysis (PC–CVA). This paper proposes an alternative approach in which discriminant analysis is carried out by a weighted principal component analysis of the group means (DPCA). This method does not require prior data reduction and produces discriminant factors that are orthogonal in the original data space.The theory and performance of the two methods are compared. Although the individual factors of DPCA are found to be less discriminating than PC–CVA, the overall discrimination, calculated by multivariate analysis of variance, and the predictive value, estimated by the leaving‐one‐out error rate, are broadly com

ISSN:0886-9383    

DOI:10.1002/cem.1180030407

出版商:John Wiley&Sons, Ltd.    

年代:1989

数据来源: WILEY

摘要:

AbstractA membrane‐discriminated gas phase analyzer is proposed for multicomponent determinations. Nitrogen gas flows countercurrent through outer and inner channels in a tube‐in‐tube arrangement. The only communication between the two channels occurs through a 500 μm aperture covered by a porous PTFE membrane. A mixture of organic compounds (up to four components) is injected into the inner channel by a heated backflushed injector and the sample components diffusing into the outer channel are monitored by a flame ionization detector (FID). A calibration set, consisting of pure components, binary, ternary and quaternary mixtures (a total of 64 samples), provides the known data base: temporal profiles of the FID output as a function of sample composition. Although the overall response behavior is not a linearly additive function of individual analyte concentrations, the use of successive inverse multiple linear regression (while continually altering the choice of the calibration samples considered for the forward regression, on the basis of the most recent values of the predicted unknown sample composition) is shown to yield analytical results for unknown samples that are in good agreement with their true

ISSN:0886-9383    

DOI:10.1002/cem.1180030408

出版商:John Wiley&Sons, Ltd.    

年代:1989

数据来源: WILEY

ISSN:0886-9383    

DOI:10.1002/cem.1180030409

出版商:John Wiley&Sons, Ltd.    

年代:1989

数据来源: WILEY

ISSN:0886-9383    

DOI:10.1002/cem.1180030402

出版商:John Wiley&Sons, Ltd.    

年代:1989

数据来源: WILEY