Santana PS

Language: Spanish
References: 3
Page: 350-361
PDF size: 545.72 Kb.

ABSTRACT

This work ends a tetralogy destinated to present the performance of several procedures that have been historically considered as solutions of methods comparison problem. By means of statistical-mathematical simulation techniques the inaccuracy incurred in the estimation of the slope of the comparison straight line, the rejection factor of the null hypothesis H_o : β = 1 (when it holds true), and the usefulness of the procedure when using indistinctively the ordinary least-square regression, the Passing-Bablok regression and the Deming regression with the λ coefficient calculated from the variance of the i-th replicate, when analytical errors were proportional and different between them, were assessed. It was assumed that the dependency of the analytical imprecision regarding the analyte concentration was monotonically increasing. None of the procedures was better than the other regarding the influence of analytical error upon estimates of the slope of the comparison straight line, the rejection factor of the null hypothesis H_o : β = 1, and the usefulness of the procedure. It is justified the search and development of personalized solutions of the methods comparison problem that incorporate the shape of the corresponding imprecision profiles within the statistical calculation machinery.

REFERENCES

1. Delgado Ramos A, Ramos Salazar R, Martínez Canalejo H, Santana Porbén S. Procederes de regresión lineal como soluciones al problema de la comparación de métodos. I. Errores analíticos constantes e iguales. RCAN Rev Cubana Aliment Nutr 2010;20:359-67. Reimpreso de: Contacto Químico [Michoacán] 2007;2(6):21-3.

2. Ramos Salazar R, Delgado Ramos A, Martínez Canalejo H, Santana Porbén S. Procederes de regresión lineal como soluciones al problema de la comparación de métodos. II. Errores analíticos constantes pero diferentes. Íbidem 2010;20:338-50. Reimpreso de: Rev Mex Patol Clín 2001;48:223-32.

3. Santana Porbén S. El problema de la comparación de métodos. El caso de los errores proporcionales e iguales. Íbidem 2013;23:179-192.