摘要
为了评价一个平稳过程的随机性,我们基于谱密度提出了一个图方法.当图中的散点呈现线性关系的时候,我们可以判定这个序列是随机的.为了说明这个思想,我们用模拟的办法来检验伪随机数的随机性.另外,我们也用了一个实际数据来考察数据的相关性.这两个例子都说明了我们的图方法是非常有效的.
To evaluate the randomness of a sequence, we propose a graphic approach based on spectral density. When the scatter points in the graph exhibit a linear structure, we conclude that the sequence is random. To illustrate the basic idea of this method, one synthetic example is carried out to assess the randomness of pseudo-random sequence generator. This method is also applied to a real data to investigate the serial correlation among a sequence. Both examples have evidenced the efficacy of our proposed graphic approach.
出处
《应用概率统计》
CSCD
北大核心
2009年第2期185-191,共7页
Chinese Journal of Applied Probability and Statistics
基金
supported by a NSF grant from National Natural Science Foundation of China (10701035)
ChenGuang project of Shanghai Education Development Foundation (2007CG33)
a special fund for young teachersin Shanghai universities (79001320)
Doctoral Program Foundation of the Ministry of Education of China(20060269016)
关键词
指数分布
周期性
随机性
序列相关
谱密度
Exponential distribution, periodogram, randomness, serial correlation, spectral density