论文部分内容阅读
为解决大数据背景下目标变量较复杂问题,本文根据格列纹科定理,提出了一种新的确定样本量方法,它是通过随机模拟样本的经验分布计算出样本量。本文讨论了该方法实现的具体步骤、优缺点等,并以指数分布为例,对规模为1万~100万的总体进行模拟。研究结果表明:(1)该方法可适用于任何形式的总体分布;(2)对于复杂的目标变量,在该方法确定的样本量下,估计量具有良好的精度;(3)该方法确定的样本量与总体分布中的参数值有关,与总体规模关系不大。
In order to solve the problem of complex objective variables in the context of big data, this paper proposes a new method to determine the sample size based on Gregorian theorem, which calculates the sample size through the empirical distribution of random simulated samples. This paper discusses the concrete steps, advantages and disadvantages of the method, and takes the exponential distribution as an example to simulate the overall scale of 1 million to 1 million. The results show that: (1) The proposed method can be applied to any form of population distribution; (2) For complex target variables, the estimator has good accuracy under the sample size determined by this method; (3) The sample size is related to the parameter values in the overall distribution and has little to do with the overall size.