Verification of Statistical Calculations in Interlaboratory Comparisons by Simulating Input Datasets

Abstract

In order to introduce a traceability chain into metrology computation, European project EMRP NEW 06 TraCIM was agreed between EC and European metrology association Euramet.One of the tasks of the project is also to establish random datasets and validation algorithms for verifying software applications for evaluating interlaboratory comparison results.Statistical analysis of different types of interlaboratory comparisons is based on numerous statistical quantities, which depend on the form of results reported by the participants, way of determining assigned value and its uncertainty, outliers etc. Complex statistical analysis is normally performed by using different software applications.In order to check the performance of those applications, we have developed validation software, which consists of a user interface, data generator and a module for calculating all standardised statistical quantities used for evaluating interlaboratory comparison results.This article is presenting our approach for verifying the validation software before offering it to the metrology community.The verification is based on simulating different cases of interlaboratory comparisons and on comparing results of statistical calculations between different SW packages like Wolfram Mathematica and Microsoft Excel.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 6 citations exist upstream.