Hybrid Bootstrap for Mapping Quantitative Trait Loci and Change PointProblems.

Hybrid Bootstrap;Mapping Quantitative Trait Loci;Change Point Problems;Statistics and Numeric Data;Science;Statistics

Sun, Ho KeunShedden, Kerby A. ;

University of Michigan

关键词: Hybrid Bootstrap;    Mapping Quantitative Trait Loci;    Change Point Problems;    Statistics and Numeric Data;    Science;    Statistics;   

Others  :  https://deepblue.lib.umich.edu/bitstream/handle/2027.42/61577/hksun_1.pdf?sequence=1&isAllowed=y

瑞士|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

The hybrid bootstrap uses resampling ideas to extend the duality approach to interval estimation for a parameter of interest when there are nuisance parameters. The confidence region constructed by the hybrid bootstrap may perform much better than the ordinary bootstrap region in situations where the data provide substantial information about the nuisance parameter, but limited information about the parameter of interest. After describing the approach, three applications will be considered. The first concerns estimating the location of a quantitative trait loci on a strand of DNA with data from a back-cross experiment. The results of some large simulation studies to demonstrate the performance of hybrid bootstrap are reported. The analysis of a real data set of rice tiller number is then presented. The second application concerns change point problems. The hybrid confidence region for a post change mean is considered after a change is detected by a Shewhart control chart in a sequence of independent normal variables. The hybrid regions are constructed in ways using likelihood ratio and Bayesian statistics. Their performance are also compared in the simulation study. The last application concerns a signal plus Poisson model of interest in high energy physics. Surprisingly, for this example the method is inconsistent--coverage probabilities to not converge to the nominal value as information about the background rate increases.