2010
DOI: 10.1016/j.nima.2009.10.156
|View full text |Cite
|
Sign up to set email alerts
|

Frequentist evaluation of intervals estimated for a binomial parameter and for the ratio of Poisson means

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
24
0

Year Published

2012
2012
2021
2021

Publication Types

Select...
5
2
1

Relationship

0
8

Authors

Journals

citations
Cited by 19 publications
(24 citation statements)
references
References 67 publications
0
24
0
Order By: Relevance
“…The signal efficiencies and the corresponding uncertainties due to limited sample sizes are calculated with the Wilson score interval [76]. Typical values of the uncertainties for SR bins with at least 5% of the yields at a given signal point are within 1-4%.…”
Section: Sourcementioning
confidence: 99%
“…The signal efficiencies and the corresponding uncertainties due to limited sample sizes are calculated with the Wilson score interval [76]. Typical values of the uncertainties for SR bins with at least 5% of the yields at a given signal point are within 1-4%.…”
Section: Sourcementioning
confidence: 99%
“…A frequentist study of the coverage for different solutions of the Poisson problem, in which the reference posterior (which corresponds to the use of the Jeffreys' prior) is also included, is presented in Ref. [17]. We adopted a similar treatment to study few examples in which the small sample size is clearly far from the asymptotic limit, such that the coverage properties are not expected to be ideal.…”
Section: E[b]mentioning
confidence: 99%
“…The horizontal line in 1D plots shows the nominal coverage. Figures 12,13,14,15,16,and 17 show the false exclusion rate (which is the average fraction of times the true signal is above the 95% upper limit) and the bias of the posterior mode, mean and median (which is the difference between the estimator and the true signal), as a function of true signal and background (left plots) and as a function of the true signal alone, for the case in which the true background coincides with the prior expectation. Only the posteriors obtained with a prior background expectation E[b] = 2 counts with relative uncertainties of 10%, 50% and 100% are shown.…”
Section: E Summaries Of the Solutions And Their Propertiesmentioning
confidence: 99%
“…[2,3,7] . In our case the technique amounts to taking only half the probability in the rst term of Eq.…”
Section: Sumerling and Darby Mid-p Versionmentioning
confidence: 99%
“…It has long been known that this result is due to the discrete nature of counting statistics and the e ects are especially severe in the low-level region (see e.g. [2,7,14]). Interestingly, the most well-known method in the health physics eld, given by Currie [9], also showed the worst result with regard to α , even for intermediate count rates, while the method of Stapleton showed good results, i.e.…”
Section: Introductionmentioning
confidence: 99%