Determining the significance of associations between two series of discrete events : bootstrap methods /

Niehof, Jonathan T. ; Morley, Steven K.

Los Alamos National Laboratory

关键词: Computers;    Statistical Mechanics;    Computer Codes;    99 General And Miscellaneous//Mathematics, Computing, And Information Science;   

DOI  :  10.2172/1035497
RP-ID  :  LA-14453
RP-ID  :  DE-AC52-06NA25396
RP-ID  :  1035497

美国|英语

来源: UNT Digital Library

【 摘 要 】

We review and develop techniques to determine associations between series of discrete events. The bootstrap, a nonparametric statistical method, allows the determination of the significance of associations with minimal assumptions about the underlying processes. We find the key requirement for this method: one of the series must be widely spaced in time to guarantee the theoretical applicability of the bootstrap. If this condition is met, the calculated significance passes a reasonableness test. We conclude with some potential future extensions and caveats on the applicability of these methods. The techniques presented have been implemented in a Python-based software toolkit.