Abstract
The basic theory of QTL (Quantitative Trait Loci) mapping is to score a population for a quantitative trait according to the marker genotype, and then to use statistics to identify differences associated with the markers and the quantitative trait of interest. Permutation based methods have been used to estimate threshold values for quantitative mapping. The permutation test based on the Student t-test for equality of means does not control Type I error rate to its nominal value when variances are unequal. In this study we propose a modification of the Student t-test based on the jackknife estimator of population variance. Janssen [20] had proposed a permutation version of the Welch test to compare equality of means under heterogeneous error distributions. The Monte Carlo method is used to compare the type I error rate of the proposed jackknife test, Janssen’s permutation test, and permutation test based on the Student t-test. The Monte Carlo study also compares the power of the proposed jackknife test, Janssen’s permutation test, and permutation test based on the Student t-test. Also, the power for each test was calculated and compared after adjusting for Type I error rates.
Keywords
Quantitative Trait Loci, Jackknife Estimator, Student t-test, Permutation test, Power, Type I error rate
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Recommended Citation
Singh, Pradeep and Rai, Shesh N.
(2010).
"ON TESTING FOR SIGNIFICANT QUANTITATIVE TRAIT LOCI (QTL) EFFECTS WHEN VARIANCES ARE UNEQUAL,"
Conference on Applied Statistics in Agriculture.
https://doi.org/10.4148/2475-7772.1066
ON TESTING FOR SIGNIFICANT QUANTITATIVE TRAIT LOCI (QTL) EFFECTS WHEN VARIANCES ARE UNEQUAL
The basic theory of QTL (Quantitative Trait Loci) mapping is to score a population for a quantitative trait according to the marker genotype, and then to use statistics to identify differences associated with the markers and the quantitative trait of interest. Permutation based methods have been used to estimate threshold values for quantitative mapping. The permutation test based on the Student t-test for equality of means does not control Type I error rate to its nominal value when variances are unequal. In this study we propose a modification of the Student t-test based on the jackknife estimator of population variance. Janssen [20] had proposed a permutation version of the Welch test to compare equality of means under heterogeneous error distributions. The Monte Carlo method is used to compare the type I error rate of the proposed jackknife test, Janssen’s permutation test, and permutation test based on the Student t-test. The Monte Carlo study also compares the power of the proposed jackknife test, Janssen’s permutation test, and permutation test based on the Student t-test. Also, the power for each test was calculated and compared after adjusting for Type I error rates.
