首页> 外文会议> >Testing Differences between Case and Control Point Patterns Using Nearest Neighbour Distances and Bootstrapping
【24h】

Testing Differences between Case and Control Point Patterns Using Nearest Neighbour Distances and Bootstrapping

机译:使用最近的邻居距离和自举来测试案例和控制点模式之间的差异

获取原文
获取原文并翻译 | 示例

摘要

This paper proposes a method for comparing point patterns in case and control data by using nearest neighbour distances, bootstrapping and the Wilcoxon Signed-Rank Test. Case-control comparisons are frequently used in medical geography and epidemiology to examine the patterns of disease and infer transmission of infectious pathogens. Our strategy addresses the problem of handling spatial analysis when the numbers of cases differ from controls. Differences in sample sizes can affect density of points and therefore bias nearest neighbour distances. To demonstrate this method we created a control set of 250 points and two sets of cases 125 points each. Bootstrapping the control data and comparing each run statistically to the cases can provide confidence intervals and estimate the risk of erroneously rejecting the null hypothesis. We follow with a case study of tuberculosis. The spatial distributions of different bacterial strains were compared and the nearest neighbour distances were analyzed as a surrogate for possible transmission of tuberculosis. The method may be useful to epidemiologists, geologists, biologists, geographers and ecologists for evaluating differences between the spatial structures of points.
机译:本文提出了一种使用最近邻距离,自举和Wilcoxon符号秩检验来比较案例数据和控制数据中的点模式的方法。病例对照比较经常在医学地理和流行病学中用于检查疾病模式和推断传染病原体的传播。当案例数量与控件不同时,我们的策略解决了处理空间分析的问题。样本大小的差异会影响点的密度,因此会使最近的邻居距离产生偏差。为了演示此方法,我们创建了一个250点的控件集和两组125点的案例集。引导控制数据并在统计上将每次运行与案例进行比较可以提供置信区间,并估计错误拒绝无效假设的风险。我们接下来以结核病为例进行研究。比较了不同细菌菌株的空间分布,并分析了最近的邻居距离,以作为可能传播结核的替代方法。该方法对于流行病学家,地质学家,生物学家,地理学家和生态学家评估点的空间结构之间的差异可能有用。

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