Validation and Error

Second- and higher-order data generation and calibration: A tutorial

“…The development of new multi-way analytical methods demands to be able to 889 estimate the corresponding analytical figures of merit, in order to compare with 890 previously existing methodologies and to report detection capabilities and other 891 important features [10]. In this context, the sensitivity is a crucial parameter, because: 892…”

Analytical Figures of Merit

“…2 We note that additional figures of merit exist beyond those discussed in this chapter. On one hand, an analytical result must always be accompanied by an uncertainty statement, because this greatly helps in interpreting the validity of the result.…”

“…Traditionally, the limit of detection (LOD) for an analyte in a sample was conceived as the minimum concentration that could be detected by the applied method. The line separating regions (1) and (2) is called decision limit, and the line separating regions (2) and (3) is called detection limit. However, modern considerations indicate that this idea is not correct, and that it is better to think about detection on a three-region basis, as depicted in Figures 6.3 and 6.4.…”

“…2 We note that additional figures of merit exist beyond those discussed in this chapter. In engineering, figures of merit are often defined for particular materials or devices in order to determine their relative utility for an application.…”

A novel methodology for determining effectiveness of preprocessing methods in reducing undesired spectral variability in near infrared spectra

“…11 Repeatability conditions include the same measurement procedure, the same operator, the same instrument and measurement conditions, the same location, and a short interval between repetitions. 12 On the other hand, reproducibility implies successive measurements of the same sample under changing measurement conditions, 13 such as measurement principle, measurement method, operator, measurement instrument, reference standard, location, conditions of use, and time. spectral acquisition is very sensitive to any change in measurement.…”