...
首页> 外文期刊>Communications in Statistics >Repeated confidence intervals and prediction intervals using stochastic curtailment under fractional Brownian motion
【24h】

Repeated confidence intervals and prediction intervals using stochastic curtailment under fractional Brownian motion

机译:在分数布朗运动下使用随机缩减的重复置信区间和预测间隔

获取原文
获取原文并翻译 | 示例
           

摘要

Repeated confidence intervals (RCIs) and prediction intervals (PIs) can be used for the design and monitoring of group sequential trials. Stochastically curtailed tests (SCTs) under fractional Brownian motion (FBM) have been studied for the interim analysis of clinical trials (Zhang etal., 2015). In this article, we derive RCIs and PIs based on SCTs under FBM for one-sided derived tests (Jennison and Turnbull, 2000). Comparisons of RCI width and sample size requirement are made to those under Brownian motion (BM) and to those of Pocock and O'Brien-Fleming design types for various type I, type II error rates, and number of interim analyses. Interim data from Beta-Blocker Heart Attack Trial are used to illustrate how to design and monitor clinical trials using these RCIs and PIs under FBM. Results show that these one-sided derived tests based on SCTs have narrower final confidence intervals and require smaller sample sizes than those using classical group sequential designs. The Hurst parameter has more impact on the RCI width than on the sample size requirements for the proposed designs.
机译:重复置信区间(RCI)和预测间隔(PIS)可用于对组顺序试验的设计和监测。已经研究了在分数布朗运动(FBM)下的随机缩减测试(SCTS)用于临床试验的临时分析(张素。,2015)。在本文中,我们基于FBM下的SCTS获得RCIS和PIS,用于片面导出测试(Jennison和Turnbull,2000)。 RCI宽度和样本尺寸要求的比较是针对布朗运动(BM)下的那些,以及用于各种I型,II型错误率的Pocock和O'Brien-Fleming设计类型和临时分析数。来自Beta-Blocker心脏病发作试验的临时数据用于说明如何在FBM下使用这些RCI和PIS设计和监测临床试验。结果表明,基于SCTS的这些单面推导测试具有较窄的最终置信区间,并且需要比使用经典群组顺序设计的样本大小更小的样本尺寸。 HURST参数对RCI宽度的影响更多,而不是所提出的设计的样本大小要求。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号