Meloun M, Hill M, Militký J, Kupka K
Department of Analytical Chemistry, Faculty of Chemical Technology, Pardubice University, Czech Republic.
Clin Chem Lab Med. 2001 Jan;39(1):53-61. doi: 10.1515/CCLM.2001.013.
Statistical software often offers a list of various descriptive statistics of location and scale, but rarely selects an efficient estimate that is statistically adequate for an actual univariate sample. The sample interval estimate for a specified degree of uncertainty seems to be more meaningful if it covers an unknown value of the population parameter. The concept of an interval estimate in medicine is then used for medical decision-making. The proposed methodology, which uses the S-Plus algorithm for biochemical, biological and clinical data analysis contains the following steps: (i) Exploratory data analysis identifies basic statistical features and patterns of the data, the distributions of which are mostly non-normal, non-homogeneous and often corrupted by outliers. (ii) Sample assumptions about data, independence of sample elements, normality and homogeneity are examined. (iii) Power transformation and the Box-Cox transformation to improve sample symmetry and stabilize the spread. (iv) Classical and robust statistics for both large (n>30) and medium-sized samples (15<n<30), point and interval estimates for the parameters of location, scale and shape. For an analysis of small samples (4<n<20) the Horn procedure of pivot measures is recommended. The proposed methodology is demonstrated in two case studies, a large sample analysis of mean pregnenolone concentrations in the umbilical blood of newborns, and a small sample analysis of mean haptoglobin concentrations in human serum.
统计软件通常会提供一系列关于位置和尺度的各种描述性统计量，但很少会选择一个对实际单变量样本在统计上足够有效的估计量。如果指定不确定度的样本区间估计涵盖了总体参数的未知值，那么它似乎更有意义。医学中的区间估计概念随后被用于医学决策。所提出的方法使用S-Plus算法进行生化、生物和临床数据分析，包括以下步骤：(i)探索性数据分析确定数据的基本统计特征和模式，其分布大多是非正态、非齐次的，并且经常受到异常值的影响。(ii)检查关于数据的样本假设、样本元素的独立性、正态性和齐次性。(iii)进行幂变换和Box-Cox变换以改善样本对称性并稳定离散程度。(iv)针对大样本(n>30)和中等样本(15<n<30)的经典和稳健统计量，对位置、尺度和形状参数进行点估计和区间估计。对于小样本(4<n<20)的分析，建议使用枢轴量的霍恩程序。在两个案例研究中展示了所提出的方法，一个是对新生儿脐血中孕烯醇酮浓度均值的大样本分析，另一个是对人血清中触珠蛋白浓度均值的小样本分析。
