No abstract

No abstract

No abstract

No abstract

No abstract

X-ray counting statistics plays a key role in establishing confidence limits in composition determination by X-ray microanalysis. The process starts with measurements of intensity on one or more samples and standards as well as related background determinations. Since each individual measurement is subject to variability associated with counting statistics, it is necessary to combine all of the counting variability according to established mathematical procedures. The next step is to apply propagation of error calculations to equations for quantitative analysis and determine confidence limits in reported composition. Similar concepts can also be applied to trace element determination. This approach can then be combined with spectral simulation modeling, making it possible to predict detectability limits without additional measurements.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.This article evaluates the accuracy of approximate confidence intervals for parameters and quantiles of the smallest extreme value and normal distributions. The findings also apply to the Weibull and the lognormal distributions. The interval estimates are based on (a) the asymptotic normality of the maximum likelihood estimator, (b) the asymptotic X2 distribution of the likelihood ratio (LR) statistic, (c) a mean and variance correction to the signed square roots of the LR statistic, and (d) the Bartlett correction to the LR statistic. The extensiveMonte Carlo results about true error probabilities and average lengths under various degrees of censoring show advantages of the LR-based intervals. For complete or moderately censored samples, the mean and variance correction to the LR statistic gives nearly exact and symmetric error probabilities. In small samples with heavy censoring, the Bartlett correction tends to give conservative error probabilities, whereas the uncorrected LR interval is often anticonservative. The results also indicate that LR-based methods have longer interval lengths than intervals based on the asymptotic normality of the maximum likelihood estimator.

No abstract