Title page for etd-0723108-160542


[Back to Results | New Search]

URN etd-0723108-160542
Author Chiang-Ming Shih
Author's Email Address No Public.
Statistics This thesis had been viewed 5101 times. Download 1340 times.
Department Applied Mathematics
Year 2007
Semester 2
Degree Master
Type of Document
Language English
Title Power Analysis of Bootstrap Methods for Testing Homogeneity of Variances with Small Sample
Date of Defense 2008-06-19
Page Count 24
Keyword
  • Levene's test
  • Bartlett's test
  • Bootstrap method
  • Abstract Several classical tests are investigated for testing the homogeneity of variances. However, in case of homoscedasticity statistics do not perform well with small sample size. In this article we discuss the use of bootstrap technique for the problem of testing equality of variances with small samples. Two important features of the proposed resampling method are their flexibility and robustness. Both α levels and power of our new proposed procedure is compared with the other classical methods discussed here.
    Advisory Committee
  • Mei-Hui Guo - chair
  • Fu-Chuen Chang - co-chair
  • Ray-Bing Chen - co-chair
  • Kam-Fai Wong - co-chair
  • Mong-Na Lo Huang - advisor
  • Files
  • etd-0723108-160542.pdf
  • indicate accessible in a year
    Date of Submission 2008-07-23

    [Back to Results | New Search]


    Browse | Search All Available ETDs

    If you have more questions or technical problems, please contact eThesys