Benchmarked linear shrinkage prediction in the Fay–Herriot small area model

The empirical best linear unbiased predictor (EBLUP) is a linear shrinkage of the direct estimate toward the regression estimate and useful for the small area estimation in the sense of increasing precision of estimation of small area means. However, one potential difficulty of EBLUP is that the overall estimate for a larger geographical area based on a sum of EBLUP is not necessarily identical to the corresponding direct estimate like the overall sample mean. To fix this problem, the paper suggests a new method for benchmarking EBLUP in the Fay–Herriot model without assuming normality of random effects and sampling errors. The resulting benchmarked empirical linear shrinkage (BELS) predictor has novelty in the sense that coefficients for benchmarking are adjusted based on the data from each area. To measure the uncertainty of BELS, the second-order unbiased estimator of the mean squared error is derived.     

个人生物识别信息的界定

生物识别信息是对自然人的生物特征进行特定技术处理所形成的信息。个人生物特征具有不变性和唯一性的特点，这使生物识别能够一劳永逸地实现对个人身份的鉴别。经过特定技术处理所形成的生物识别信息也具有这些特点，因此任何对于生物识别信息的攻击都可能对个人产生不可逆转的影响。现实生活中，商业化收集的泛滥、限制行动自由、售卖和窃取、深度伪造以及诈骗已经成为侵害生物识别信息的重要表现。保护个人生物识别信息安全迫在眉睫，而准确界定生物识别信息是首要前提。综观国外立法经验，生物识别信息的构成应当聚焦三个要素：个人生物特征之反映、特定技术处理之环节、唯一性识别之功能。此三项要素也是认定生物识别信息的关键。生物识别信息可以通过含有个人生物特征的图像来采集，但是该图像仅仅是个人生物特征的反映。在没有进行特定技术处理之前，图像本身并不构成生物识别信息，无论该图像是电子图像抑或纸质照片。生物识别信息的本质是对个人生物特征进行的测量，即所谓的“特定技术处理”,至于测量的方式则在所不问。无论进行面对面的真人测量抑或从含有个人生物特征的图像中测量，所得的信息都是生物识别信息。单纯对于个人生物特征的描述不构成生物识别信息，个...     

On the scientific study of small samples: Challenges confronting quantitative and qualitative methodologies

Often phenomena that are important to understand and predict are very rare. Rare events can prove difficult to analyze systematically because they do not generate many sampling observations. In this article I examine how small sample sizes can be studied scientifically. The article begins with an explanation of the distinction between research and science. I then bring to the fore the importance of counterfactual comparisons and outline the nature of the methodological problems posed by the study of small samples. These problems include challenges related to using a single case, small sample sizes, selecting on the dependent variable, regression toward the mean, explaining a variable with a constant, and using the same data to both generate and test hypotheses. I provide potential resolutions to these problems which are: (a) employing matched controls; (b) shifting or widen the category of inquiry; (c) selecting variables based on variance in the independent variable; (d) including counterfactuals; (e) ensuring that both independent and dependent variables demonstrate variation; and (f) testing potential hypotheses against data sets that are fully independent of those used to generate the hypotheses. I conclude with a discussion of future directions for undertaking a more scientific approach to using small samples.