论文部分内容阅读
基于2009年8月3日11月14日北京市甲型H1N1病例的阳性率抽样数据,计算得到了后验精度估计.实验结果表明,在置信水平1-α=0.95的条件下,北京市甲型H1N1阳性率抽样检测的绝对误差均没有超过0.1,处于可控水平.甲流疫情早期,抽样检测的相对误差较大,37周以前的相对误差大于0.5.随着疫情发展,相对误差逐步降低,估计结果趋于可靠,42周后的相对误差小于0.2.最后,针对阳性病例样本检测的问题,提出了基于精度控制的最优样本量控制策略,探索了不同条件下边际样本量的变化规律,分析了阳性率与绝对误差组合控制下的最优样本量估计.
The posterior precision estimation was calculated based on the positive rate sampling data of A / H1N1 cases in Beijing on August 3, 2009 from November 14, 2009. The experimental results show that under the confidence level of 1-α = 0.95, H1N1 positive rate of sampling and testing of the absolute error were not more than 0.1, at a controlled level.In the early epidemic of influenza A, sampling relatively large relative error, 37 weeks ago, the relative error is greater than 0.5.As the epidemic develops, the relative error is gradually reduced , The estimated results tend to be reliable, and the relative error after 42 weeks is less than 0.2. Finally, based on the detection of positive case samples, an optimal sample size control strategy based on precision control is proposed, and the variation of marginal sample size under different conditions is explored , The optimal sample size under the combination of positive rate and absolute error was analyzed.