POWRE: Statistcs: Research, Education and Application

POWRE：统计：研究、教育和应用

• 批准号: 9720490
• 负责人: Deborah Nolan
• 金额: $ 7.5万
• 依托单位: University of California-Berkeley
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 1998
• 资助国家: 美国
• 起止时间: 1998-01-01 至 2000-06-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9720490&HistoricalAwards=false
• 关键词: POWRE; Statistcs; Research; Education; Application

This work is funded through the Professional Opportunities for Women in Research and Education (POWRE) program as a Research/Educational Enhancement Project. The investigator proposes a research and education plan that: draws on the wealth of interdisciplinary applications of statistics; contributes to advances in model selection and resampling methodologies; and utilizes the new capabilities for developing computing environments and courseware. Statistical applications are making greater use of flexible, computationally--intensive modeling procedures. As a result, the issue of selecting appropriate models is an important one. The investigator plans to extend methods of model selection to robust regression, including least absolute deviation regression. These alternatives to least squares regression are very popular when the data come from heavy tailed distributions and when there are outliers. In the area of education, the investigator plans to integrate statistical theory and practice in a way not commonly encountered in undergraduate courses -- through in-depth case studies or labs. The labs raise interesting scientific questions that highlight the interdisciplinary framework in which statistics has developed; they employ new statistical methodologies, such as resampling methods and model selection techniques; and they are to be presented in an interactive programming environment. The investigator also plans to bring her technical skills in high--dimensional modeling to bear on an area of statistical application -- modeling growth of roots in the study of plants. Biologists are interested in learning about the effects of various environmental conditions on biosynthesis rates in growing plant tissues. Developing a model for the velocity of growth and for the elemental growth rate using innovative nonparametric methodology would be very helpful in analyzing experimental outcomes.

这项工作是通过妇女在研究和教育领域的职业机会方案作为一个研究/教育促进项目资助的。 调查员提出了一个研究和教育计划，即：利用统计学的跨学科应用的财富;有助于模型选择和重新排序方法的进步;并利用开发计算环境和课件的新功能。 统计应用程序正在更多地利用灵活的、计算密集型的建模过程。因此，选择适当的模式是一个重要的问题。研究者计划将模型选择方法扩展到稳健回归，包括最小绝对偏差回归。当数据来自重尾分布和存在离群值时，这些替代最小二乘回归的方法非常流行。 在教育领域，研究人员计划通过深入的案例研究或实验室，以本科课程中不常见的方式整合统计理论和实践。这些实验室提出了有趣的科学问题，突出了统计学发展的跨学科框架;它们采用了新的统计方法，如恢复方法和模型选择技术;它们将在交互式编程环境中提出。 调查员还计划把她在高维建模方面的技术技能应用于统计应用领域--植物研究中的根系生长建模。 生物学家对了解各种环境条件对生长中的植物组织的生物合成速率的影响很感兴趣。 发展一个模型的增长速度和元素的增长率使用创新的非参数方法将是非常有帮助的分析实验结果。