Testing of data evaluation methods

Kafala, S. I.; MacMahon, T. D.; Gray, P.W.

doi:10.1016/0168-9002(94)91796-5

Cited by 25 publications

(9 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…3) Our best value and its standard deviation, calculated with the BM using the full data sets (except data omitted by Chechev and Egorov, that were also omitted here), are shown in Table 2, as well as the recommended results presented in those papers. 1,3) Comparing columns 2 and 3 of Table 2, we observe that the BM gives uncertainties greater than the method of Kafala et a1. When comparing the BM with the Chechev and Egorov method, we observe that in three cases (Tritium, 35S and 55Fe), the differences between both methods are not very large.…”

Section: Applying the Bootstrap Methodsmentioning

confidence: 80%

“…These quantities behave very much like the usual average and its standard deviation. The bootstrap method presented here is the method with the simplest set of requirements from all usual methods of dealing with discrepant data set, since it requires only the unbiasedness of the data; the other methods [1][2][3][4][5] require the unbiasedness and other hypothesis. Moreover, the existence of one strongly biased datum (or a few data in a large data set) does not affect the BM estimate as was shown in section II.3., in contrast to many other methods.…”

Section: Discussionmentioning

confidence: 99%

“…Besides having a direct effect on the variance, we must consider the displacement of the median by the outlier when it does not become the median. We are now considering the data randomly distributed around the true value, and use its average as a reference value 1 which replaces x in formula (6). In this case of a single outlier, the median is pulled by the outlier about half of one place in the ordered data set in average.…”

Section: Dependence On Outliersmentioning

confidence: 99%

“…[1][2][3][4][5] All the proposed procedures, however, assume that incorrect uncertainties are responsible for the data discrepancy, and modify them usually by a common factor, keeping unchanged the central values.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Characteristics of the Bootstrap Estimate of Discrepant Data Sets

Helene¹,

Vanin²,

Castro³

2002

Journal of Nuclear Science and Technology

View full text Add to dashboard Cite

We present in this paper a Bootstrap Method to estimate a best value and its standard deviation for a discrepant set of data. The method is applied to the determination of the half-lives of 137 Cs, 90Sr, 252Cf, Tritium, 35 S, 55Fe, and 99Mo. The self-consistency of the method and its capability to reject outliers was tested with good results. The application of the method to the determination of covariance matrices was tried, with poor results.

show abstract

Section: Applying the Bootstrap Methodsmentioning

confidence: 80%

Section: Discussionmentioning

confidence: 99%

Section: Dependence On Outliersmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Characteristics of the Bootstrap Estimate of Discrepant Data Sets

Helene¹,

Vanin²,

Castro³

2002

Journal of Nuclear Science and Technology

View full text Add to dashboard Cite

show abstract

“…Many other methods have been suggested for treatment of inconsistent data and Ref. [8] compares the performance of some different methods. In [8], the Bayesian procedure to adjust uncertainties performed best.…”

Section: Introductionmentioning

confidence: 99%

Monte Carlo integral adjustment of nuclear data libraries – experimental covariances and inconsistent data

Sjöstrand

Schnabel

2019

EPJ Web Conf.

View full text Add to dashboard Cite

Integral experiments can be used to adjust nuclear data libraries. Here a Bayesian Monte Carlo method based on assigning weights to the different random files is used. If the experiments are inconsistent within them-self or with the nuclear data it is shown that the adjustment procedure can lead to undesirable results. Therefore, a technique to treat inconsistent data is presented. The technique is based on the optimization of the marginal likelihood which is approximated by a sample of model calculations. The sources to the inconsistencies are discussed and the importance to consider correlation between the different experiments is emphasized. It is found that the technique can address inconsistencies in a desirable way.

show abstract

Evaluation of decay and radiation characteristics of93mNb,111In, and170Tm

Chechev¹,

Kuzmenko²

2000

Meas Tech

View full text Add to dashboard Cite

New estimates of the decay and radiation characteristics of 93mNb, 111[n ' and 170Tin that take into account data published up to 1999 are presented. According to the research coordination program organized by the Nuclear Data Division of the International Atomic Energy Agency [ 1], the recommended data on the decay of nuclides used as X-ray and "~-ray standards [2] are to be revised and expanded in 1999-2000.Within the framework of this program, in which the authors' participation is supported by a research contract (RC 10315) with the International Atomic Energy Agency, we evaluated the decay and radiation characteristics of 12 radionuclides [3]. Here we present new estimates lor three of them: 93mNb, I I lln ' and 17~ The estimates are based on experimental data published in the world physics literature up to 1999. 1, 93mNb. The most practically important nuclear-physics characteristics (NPC) of 93mNb -the half-life and the absolute intensity of the KX components -were evaluated in [2] from the data available in 1991. The recommended data [2] were also used in a recent list of 93mNb NPCs in Nuclear Data Sheets [4]. In the data available in 1999, as compared with the initial experimental lbr the estimates of [2, 4], we found only one additional experimental result associated with indirect measurement of the absolute intensity of the Kc~ radiation of Nb by comparing measurements of the specific activity of a 93mNb solution by two methods: using a liquid scintillator and a Si(Li) detector [5]. This result confirmed a previous measurement of the KX radiation of Nb by the same Si(Li) detector [6]. Here, unlike in [4], we use as reference quantities to estimate the NPCs of 93mNb the total absolute intensity of KX radiation of Nb and the theoretical value of the internal conversion coefficient (ICC) of the 30.77-keV gamma transition (cq< = 2.62(8)-104) obtained for multipole order M4 by computer interpolation [7] from ICC tables [8]. The relative error of cc K is assumed to be 3%, in accordance with available estimates of the reliability of calculations of theoretical ICC values with a pure multipole composition. The adopted theoretical value ~Kt is in good agreement with the experimental ~Ke = 2"58( 15)'104 4 5 [9] and differs from ~Ke = 1.7(3)-10 calculated in [10] from the measured ratio Ie/IKx = 8(I)-10-'. 93mThe obtained estimates of the decay and radiation characteristics of Nb are listed in Tables i and 2. We estimated the half-life of 93mNb earlier [11][12][13]. A list of the experimental data used to estimate Tl~ can be found in [12]. The procedures for estimation and statistical processing of data sets with various degrees of divergence are described in [13].For gamma radiation with energy E~, = 30.77(2) keV and multipole order M4, the internal conversion coefficients were ~K = 2-62( 8)'104, ~L = 1-216(40)'105, 0~M = 2"69(8)'104, and ~tot = 1-79(5) "105-The total absolute intensity of KX radiation -~-/KX was estimated as the weighted mean of five measurement results. They included 11.0(4)% calculated us...

show abstract

Testing of data evaluation methods

Cited by 25 publications

References 5 publications

Characteristics of the Bootstrap Estimate of Discrepant Data Sets

Characteristics of the Bootstrap Estimate of Discrepant Data Sets

Monte Carlo integral adjustment of nuclear data libraries – experimental covariances and inconsistent data

Evaluation of decay and radiation characteristics of93mNb,111In, and170Tm

Contact Info

Product

Resources

About