Two-Sample Statistical Testing for Weighted Data Sets

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21340%2F17%3A00316346" target="_blank" >RIV/68407700:21340/17:00316346 - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—

Result language
angličtina
Original language name
Two-Sample Statistical Testing for Weighted Data Sets
Original language description
We introduce a field-specific use case for data weighting in high energy physics. This weighting procedure naturally influences the subsequent applications of machine learning methods which require prior statistical homogeneity testing. We propose modifications of various homogeneity tests directly applicable to the weighted data sets. As a mutual validation we are using tests based on an empirical distribution function compared to tests using phi-divergence measures. Weighted modifications of classical homogeneity tests are verified by numerical simulation for several principal univariate probability distributions. Also, we present a specific rearranging procedure which transforms the weighted observations into unweighted data sets so that it enables usage of standard unweighted tests.
Czech name
—
Czech description
—

Project
Result was created during the realization of more than one project. More information in the Projects tab.
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach

Publication year
2017
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Article name in the collection
SPMS 2017 - Stochastic and Physical Monitoring Systems, Proceedings of the international conference
ISBN
978-80-01-06338-5
ISSN
—
e-ISSN
—
Number of pages
9
Pages from-to
1-9
Publisher name
Česká technika - nakladatelství ČVUT
Place of publication
Praha
Event location
Dobřichovice
Event date
Jun 19, 2017
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
000425554500001

Similar results(10)