Issue 5. 2023
-
Full publication
-
Doc. Document
-
- Presentation of Volume 5, 1, 2023 José María Sarabia
-
Audit sampling as a quality standard for
multisource official statistics
Li-Chun Zhang
-
Doc. - DOI
- https://doi.org/10.37830/SJS.2023.1.05
- Abstract
Designed surveys through sampling or census are the standard approach to official statistics, where the targets are descriptive summaries of a given population. Official statistics are also commonly produced by combining relevant administrative registers, such as in the Nordic countries since the 1960s. The scope of non-survey data sources are being extended to include various so-called big-data sources, although so far relatively few multisource statistics of this kind have been credited as official statistics. Trustworthy evaluation of multisource official statistics is a fundamental issue for creating a new quality assurance standard. In this paper, audit sampling inference will be explained, illustrated and promoted to this end.
- Keywords
- Descriptive inference, survey sampling, auditing, census, social media index, register household
-
-
Applying and Testing Benford¿s Law Are Not the
Same
William M. Goodman
-
Doc. - DOI
- https://doi.org/10.37830/SJS.2023.1.03
- Abstract
Most papers on Benford¿s Law primarily discuss either (1) the science and mathematics for explaining the law; or (2) how to apply the law, especially for detecting data manipulation and fraud; or (3) suggestions for statistical tests to determine if data conform to a Benford¿s distribution. Leonardo Campanelli¿s recent paper ¿Testing Benford¿s Law¿ strongly objects to a descriptive measure I discussed in my paper ¿The Promises and Pitfalls of Benford¿s Law¿¿as if that measure were intended for Benford¿s testing in the Category-3 sense relevant for Campanelli¿s paper (SJS, vol. 4, 2022). This reflects a conflation of meanings for ¿testing¿ that is common in the Benford¿s literature, where many Category-2 papers claim they are applying (directly) conventional or new hypothesis tests as tools to detect fraud. Yet, fraud detection is a forensic and context-sensitive process, for which there is no set formula. In this paper, I clarify the sampling plan I had used earlier to collect and analyze a quasi-random sample of datasets, based on published criteria in the literature, to paint a tentative picture of how far real data vary, and in what ways, from abstract BL expectations. Further, I discuss simulations I have conducted to replicate and expand on my previous results.
- Keywords
- Benford¿s Law Applications, Benford¿s-Law Conformance Tests, Fraud detection, Benford¿s-Law Error Distributions
-
-
A generalization of the transmuted Rayleigh
distribution
Hugo S. Salinas, Guillermo Martínez-Flórez, Yuri A. Iriarte, Artur J. Lemonte
-
Doc. - DOI
- https://doi.org/10.37830/SJS.2023.1.04
- Abstract
In this paper, we introduce a new family of distributions for modeling positive data. The new distribution arises from the quotient of two independent random variables: transmuted Rayleigh in the numerator, and beta in the denominator. Structural properties of the new distribution are derived, and an application to real data reveals good performance of this new distribution in practice.
- Keywords
- Kurtosis, Rayleigh distribution, Slash distribution, Skewness, Transmuted distributions
-
-
Speech at the National Statistics. Award 2022
Ceremony
Enrique Castillo
-
Doc. - DOI
- https://doi.org/10.37830/SJS.2023.1.02
- Abstract
This work presents the speech given at the 2022 National Statistics Award ceremony, summarizing the main scientific contributions of the recipient, Professor Enrique Castillo. These contributions include significant advancements in the field of extreme value statistics, where he provides analytical and graphical methods for identifying tail types, conditional specification, Bayesian networks, addressing compatibility issues, sensitivity analyses in optimization problems with closedform solutions, solving linear systems of inequalities, demonstrating that polytopes are the unique bounded solutions, fatigue models based on properties, especially the S-N and crack growth models, which provide the only models satisfying certain necessary compatibility conditions. Additionally, it covers probabilistic safety analyses of nuclear power plants, roads, and railways, allowing the assessment of risks using statistical models that include thousands of variables, as well as applications in artificial intelligence and Bayesian methods that expand the range of possible solutions by considering mixtures of much more limited distribution families.
- Keywords
- Extreme values, conditional distribution specification, Bayesian networks, operations research, fatigue models, nuclear safety, road and railway safety, artificial intelligence, Bayesian methods
-