INSTRUMENTAL ANALYSIS

仪器分析

• 批准号: 3896391
• 负责人: G H WEISS
• 金额: --
• 依托单位: CENTER FOR INFORMATION TECHNOLOGY
• 依托单位国家: 美国
• 资助国家: 美国
• 起止时间: 至
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/3896391
• 关键词: computer data analysis; density; interferometry; mathematical model; monograph; nuclear magnetic resonance spectroscopy; statistics /biometry

Two projects are included under this heading:1. The estimation and minimization of error in the design cf NMR experiments for chemical analysis, and 2. The calculation of exact representations of probability density functions required for the use of intensity statistics and direct methods for the interpretation of crystallographic data. In the first of these we have analyzed the bias introduced in the estimation of the area under a single peak by the use of apodization filtering. If one does not average over the range of possible bias, but rather considers the worst possible bias induced by an apodizing filter then it is shown that the error induced by the bias can reach 100% or more, depending on the discretizing intervals. In the area of crystallography we have developed the theory required to calculate the probability density functions for crystals with non-symmetric distributions of structure-factor magnitudes. In addition we have begun to examine a variety of possible techniques to correct measurements of low counts which are of the order of magnitude of background noise. The present method simply subtracts the average background which can result in negative counts, which are unphysical. A further part of the project is the writing of an introductory monograph on crystallographic statistics.

本标题下包括两个项目：1.的估计和 化学核磁共振实验设计中的误差最小化 分析，2。概率精确表示的计算 密度函数所需的使用强度统计和直接 解释晶体学数据的方法。 在第一部分中，我们分析了 用切趾法估计单峰下的面积 过滤如果一个人不平均的范围内可能的偏见，但 而是考虑由切趾滤波器引起的最坏可能偏差 然后示出了由偏置引起的误差可以达到100%或更大， 这取决于离散化间隔。 在晶体学领域，我们已经发展了所需的理论， 计算非对称晶体的概率密度函数 结构因子幅值的分布。此外，我们已开始 检查各种可能的技术，以纠正测量低 数量级为背景噪声的计数。本 方法简单地减去平均背景，这可能导致负 计数，这是非物理的。该项目的另一部分是写作 结晶统计学的入门专著