Nonlinear statistical inference and boundary crossing probabilities

非线性统计推断和边界交叉概率

• 批准号: 227197-2009
• 负责人: Wang, Liqun
• 金额: $ 2.11万
• 依托单位: University of Manitoba
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2014
• 资助国家: 加拿大
• 起止时间: 2014-01-01 至 2015-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=546719
• 关键词: Nonlinear; statistical; inference; boundary; crossing

The proposed research is concerned with finding relationships between a response variable and other predictor variables, some of which are either not directly observable or measured with error. Examples include personal exposure to air pollutants, long-term systolic blood pressure, social ability and wealth. It is known that statistical methods ignoring the measurement error result in biased and inconsistent estimates. The goal of the proposed research is to develop novel methods to accurately estimate relationships in the presence of measurement error, which do not rely on restrictive assumptions and are therefore applicable to a wide range of problems. The anticipated results will advance statistical theory and methodology, as well as provide better tools for real data analysis.

拟议的研究涉及寻找响应变量和其他预测变量之间的关系，其中一些变量要么无法直接观察，要么测量有误差。例子包括个人接触空气污染物、长期收缩压、社交能力和财富。众所周知，忽略测量误差的统计方法会导致估计有偏差和不一致。所提出的研究的目标是开发新的方法来准确估计存在测量误差的关系，该方法不依赖于限制性假设，因此适用于广泛的问题。预期结果将推进统计理论和方法，并为实际数据分析提供更好的工具。