Department of Mathematics, The College of William & Mary, Williamsburg, Virginia.
Stat Med. 2019 Aug 15;38(18):3460-3475. doi: 10.1002/sim.8189. Epub 2019 May 17.
We propose two measures of performance for a confidence interval for a binomial proportion p: the root mean squared error and the mean absolute deviation. We also devise a confidence interval for p based on the actual coverage function that combines several existing approximate confidence intervals. This "Ensemble" confidence interval has improved statistical properties over the constituent confidence intervals. Software in an R package, which can be used in devising and assessing these confidence intervals, is available on CRAN.
我们提出了两种用于二项式比例 p 的置信区间的性能度量:均方根误差和平均绝对偏差。我们还基于实际覆盖函数设计了一个基于实际覆盖函数的 p 置信区间,该函数结合了几个现有的近似置信区间。这个“合奏”置信区间在统计性质上优于组成的置信区间。可以在设计和评估这些置信区间时使用的 R 包中的软件可在 CRAN 上获得。