Premium
Estimation of bias for single‐point calibration
Author(s) -
Bánfai Balázs,
Kemény Sándor
Publication year - 2012
Publication title -
journal of chemometrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.47
H-Index - 92
eISSN - 1099-128X
pISSN - 0886-9383
DOI - 10.1002/cem.2417
Subject(s) - confidence interval , statistics , calibration , mathematics , point estimation , interval estimation , zero (linguistics) , point (geometry) , sample size determination , interval (graph theory) , econometrics , algorithm , combinatorics , philosophy , linguistics , geometry
The applicability of single‐point calibration is usually decided by testing the hypothesis of zero intercept. It is flawed, first, because the zero intercept does not infer that it is true, and second, accepting a zero intercept does not imply a low bias in the estimated concentration of the unknown sample. Instead, a method is proposed based on the two one‐sided tests (interval hypothesis) for the bias. The test is performed by comparing a confidence interval for the bias to an allowable limit, defined in concentration units. Fieller's theorem was used for the ratio of two normally distributed random variables to construct the confidence interval for the bias. The proposed method is appropriate for judging the extent of bias in case of single‐point calibration. Copyright © 2012 John Wiley & Sons, Ltd.