Efficient Designs for Scientific Studies with Repeated Measurements Data

PI将研究和探索在空间和/或时间上对所有实验单元进行重复测量的科学情况。FDA建议在临床试验的一个或多个阶段进行药代动力学和药效学（PK/PD）评价，或在仿制药的监管决策中进行生物利用度研究。同样地，这样的实验对于流行病学、社交网络和农业现场实验（包括营销、实体店和虚拟店）的研究也是高度期望的。PI将开发即用型统计/计算软件，这些软件将被设计、测试和免费分发，以传播有关重复测量研究的知识。研究成果将作为一个平台，以广泛的多学科研究人员和从业人员。功能性磁共振成像（fMRI）、药学、医疗器械和仿制药领域的科学家将获得非常有用的工具来开展他们的研究，并根据重复测量收集的数据进行适当和谨慎的循证推理。在同一实验单元上重复测量，也称为交叉设计，通常会产生残差/相邻/相关效应，需要对其进行建模和探索。PI将重点关注数据收集（设计）和建模方面的科学情况，这些情况以前没有得到充分解决和研究。在此过程中，将通过对理论问题的深入研究开发计算工具。健全的创新技术将被开发用于识别具有成本效益的设计，以确定最佳和接近最佳的估计的主要和，如果需要的话，次要参数嵌入在模型下重复observation.This奖项反映了NSF的法定使命，并已被认为是值得通过评估使用基金会的智力价值和更广泛的影响审查标准的支持。

The PI will study and explore scientific situations dealing with repeated measurements on all experimental units in space and or time. Such studies are recommended by FDA for pharmacokinetic and pharmacodynamics (PK/PD) evaluations in one or more phases of clinical trials or for bio-availability studies in making regulatory decision for generic drugs. Likewise, such experiments are also highly desirable for studies in epidemiology, social networks, and agricultural field experiments, including marketing, physical and virtual stores. The PI will develop ready-to-use statistical/computational soft wares which will be designed, tested, and freely distributed for dissemination of knowledge on repeated measurements studies. The research outcomes will serve as a platform to a broad class of multidisciplinary researchers and practitioners. Scientists in the fields of functional Magnetic Resonance Imaging (fMRI), pharmacy, medical devices, and generic medicine will be provided with highly useful tools to carry out their studies and make proper and prudent evidence based inference from the collected data under repeated measurements.Repeated measurements on the same experimental unit, also known as crossover designs, often produce residual/neighbor/correlated effects which need to be modeled and explored. The PI will focus on the scientific situations both in terms of data collection (design) and modeling which have not been adequately addressed and studied before. In the process, computational tools will be developed with thorough study of the theoretical issues. Sound innovative techniques will be developed for identifying cost-effective designs in order to ascertain optimal and nearly optimal estimation of the primary and, if needed, secondary parameters embedded in models under repeated observations.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.