REDUCING BIAS IN THE ANALYSIS OF COUNTING STATISTICS DATA

Authors
HAMMERSLEY AP ANTONIADIS A
Citation
Ap. Hammersley et A. Antoniadis, REDUCING BIAS IN THE ANALYSIS OF COUNTING STATISTICS DATA, Nuclear instruments & methods in physics research. Section A, Accelerators, spectrometers, detectors and associated equipment, 394(1-2), 1997, pp. 219-224
Citations number
4
Categorie Soggetti
Nuclear Sciences & Tecnology","Physics, Particles & Fields","Instument & Instrumentation",Spectroscopy
Journal title
Nuclear instruments & methods in physics research. Section A, Accelerators, spectrometers, detectors and associated equipment → ACNP
ISSN journal
01689002
Volume
394
Issue
1-2
Year of publication
1997
Pages
219 - 224
Database
ISI
SICI code
0168-9002(1997)394:1-2<219:RBITAO>2.0.ZU;2-H
Abstract
In the analysis of counting statistics data it is common practice to e stimate the variance of the measured data points as the data points th emselves. This practice introduces a bias into the results of further analysis which may be significant, and under certain circumstances lea d to false conclusions. In the case of normal weighted least squares f itting this bias is quantified and methods to avoid it are proposed.