2017/06/30 by Jan Kieseler · 1 citation
Agricultural and Biological Sciences · Decision Sciences · Mathematics · Physics and Astronomy · #Covariance #Covariance matrix #Data set #Differential (mechanical device) #Maximization #Pesticide Residue Analysis and Safety #Radioactive Decay and Measurement Techniques #Scientific Measurement and Uncertainty Evaluation #Set (abstract data type) #Simple (philosophy) #Software #hep-ex #physics.data-an #stat.AP
paper · pdf · doi:10.1140/epjc/s10052-017-5345-0
published as Eur. Phys. J. C (2017) 77: 792 · 12 pages, 15 figures (v3: changed figure ranges) Accepted by EPJC
openalex created_date 2017/06/15 · arxiv created 2017/10/27 · openalex publication_date 2017/11/01 · arxiv updated 2018/01/09 · openalex updated_date 2026/08/05
A method is discussed that allows combining sets of differential or inclusive measurements. It is assumed that at least one measurement was obtained with simultaneously fitting a set of nuisance parameters, representing sources of systematic uncertainties. As a result of beneficial constraints from the data all such fitted parameters are correlated among each other. The best approach for a combination of these measurements would be the maximization of a combined likelihood, for which the full fit model of each measurement and the original data are required. However, only in rare cases this information is publicly available. In absence of this information most commonly used combination methods are not able to account for these correlations between uncertainties, which can lead to severe biases as shown in this article. The method discussed here provides a solution for this problem. It relies on the public result and its covariance or Hessian, only, and is validated against the combined-likelihood approach. A dedicated software package implementing this method is also presented. It provides a text-based user interface alongside a C++ interface. The latter also interfaces to ROOT classes for simple combination of binned measurements such as differential cross sections.