Approximations for the tail probabilities and moments of scan statistic英文文献.pdfVIP

下载本文档

3
0
约3.3万字
约 8页
2018-02-05 发布于四川
举报
版权申诉

Approximations for the tail probabilities and moments of scan statistic英文文献.pdf

1、原创力文档（book118）网站文档一经付费（服务费），不意味着购买了该文档的版权，仅供个人/单位学习、研究之用，不得用于商业用途，未经授权，严禁复制、发行、汇编、翻译或者网络传播等，侵权必究。。
2、本站所有内容均由合作方或网友上传，本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺！文档内容仅供研究参考，付费前请自行鉴别。如您付费，意味着您自己接受本站规则且自行承担风险，本站不退款、不进行额外附加服务；查看《如何避免下载的几个坑》。如果您已付费下载过本站文档，您可以点击这里二次下载。
3、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等，请点击“版权申诉”（推荐），也可以打举报电话：400-050-0827(电话支持时间：9:00-18:30)。
4、该文档为VIP文档，如果想要下载，成为VIP会员后，下载免费。
5、成为VIP后，下载本文档将扣除1次下载权益。下载后，不支持退款、换文档。如有疑问请联系我们。
6、成为VIP后，您将拥有八大权益，权益包括：VIP文档下载权益、阅读免打扰、文档格式转换、高级专利检索、专属身份标志、高级客服、多端互通、版权登记。
7、VIP文档为合作方或网友上传，每下载1次，网站将根据用户上传文档的质量评分、类型等，对文档贡献者给予高额补贴、流量扶持。如果你也想贡献VIP文档。上传文档

STATISTICS IN MEDICINE, VOL. 12, 1845-1852 (1993) APPROXIMATIONS FOR THE TAIL PROBABILITIES AND MOMENTS OF THE SCAN STATISTIC JOSEPH GLAZ Department of Statisiics, University of Connecticut. Siorrs. Connecticut 06269-3120, U.S.A. SUMMARY The scan statistic is used to test the hypothesis that the observed events occur at random (uniformly distributed) in time or space versus the hypothesis that they cluster within a moving window of size w. To implement the testing procedure based on the scan statistic its tail probabilities have to be effectively evaluated.In this article a survey of results on the approximations of the distribution of the scan statistic and its moments is presented. Numerical results comparing these approximations are also given. Numerous references with applications in epidemiologicalstudies using the scan statistics are mentioned. Related scan statisticsthat have been used in many other interesting applications are listed in the reference section. The article concludes with the presentation of unsolved problems related to the scan statistic. 1 . INTRODUCTION Suppose that N events occur in the interval (0, TI, where the units here are measuring time or distance. It is assumed that the observed events are distributed independently and identically from a continuous distribution on the interval (0, T] and that N is fixed and known. We are interested in testing the null hypothesis that the events occur at random in (0, T ] (uniform distribution) against the alternative hypothesis that the events cluster in the interval (0,TI. - 4 In what follows, without loss of generality, we assume that T = 1