Title :
Compatibility Test Method for the Very Few Sample Based on Variable-Metric Correlation Statistics in Normal Distribution
Author :
Bin, Shan ; Wang, Yuegang ; Guo, Zhibin ; Bo, Yang
Author_Institution :
Sect. 304, Xi´´an Res. Inst. Of High-Tech., Xi´´an, China
Abstract :
It is hard to do statistical analysis of the aircraft structural response telemetry signals since the data that could be achieved is very few. It is thus required to combine the current information and the prior information. While, at the same time it provokes the problem of the very few sample compatibility between different kinds of information. This article, therefore, after the analysis of problems of the present few sample compatibility test methods in normal distribution, proposes the compatibility test method of variable-metric correlation statistics Kijk with no unknown parameters in the distribution. And this method also solves the problem of simulation calculation for the test critical value in Matlab. Through the practical application, it is proved that the method can effectively solve the problem of the very few sample compatibility test in normal distribution.
Keywords :
correlation methods; normal distribution; compatibility test method; normal distribution; simulation calculation; variable-metric correlation statistics; Aircraft; Correlation; Distribution functions; Gaussian distribution; Joints; Presses; Telemetry; compatibility test; normal distribution; variable-metric correlation statistics; very few sample;
Conference_Titel :
Computational and Information Sciences (ICCIS), 2010 International Conference on
Conference_Location :
Chengdu
Print_ISBN :
978-1-4244-8814-8
Electronic_ISBN :
978-0-7695-4270-6
DOI :
10.1109/ICCIS.2010.255