分数布朗运动与长期临床试验招募
Fractional Brownian motion and long term clinical trial recruitment.
作者信息
Zhang Qiang, Lai Dejian
机构信息
American College of Radiology, 1818 Market Street, Suite 1600, Philadelphia, PA, 19103, U.S.A.
School of Public Health, University of Texas at Houston, Division of Biostatistics, Houston, TX, 77030, USA.
出版信息
J Stat Plan Inference. 2011 May 1;141(5):1783-1788. doi: 10.1016/j.jspi.2010.11.028.
Prediction of recruitment in clinical trials has been a challenging task. Many methods have been studied, including models based on Poisson process and its large sample approximation by Brownian motion (BM), however, when the independent incremental structure is violated for BM model, we could use fractional Brownian motion to model and approximate the underlying Poisson processes with random rates. In this paper, fractional Brownian motion (FBM) is considered for such conditions and compared to BM model with illustrated examples from different trials and simulations.
在临床试验中预测入组情况一直是一项具有挑战性的任务。人们已经研究了许多方法，包括基于泊松过程及其由布朗运动（BM）进行的大样本近似的模型，然而，当布朗运动模型的独立增量结构被违反时，我们可以使用分数布朗运动来对具有随机速率的潜在泊松过程进行建模和近似。在本文中，针对此类情况考虑了分数布朗运动（FBM），并通过来自不同试验和模拟的示例与布朗运动模型进行了比较。
Zotero 插件