A new jackknife method for unbalanced variance component problems with applications in quantitative genetics /
A new jackknife method is proposed for variance estimation of a heritability estimator in unbalanced random-effects models. This work is motivated by the need for analyzing a data set obtained in a project to study the quantitative genetics of fire ants. The new method uses the concept of weighted...
| Main Author: | |
|---|---|
| Format: | Thesis Book |
| Language: | English |
| Published: |
[Place of publication not identified] :
[publisher not identified] ;
2001.
|
| Subjects: | |
| Online Access: | http://proxy.library.tamu.edu/login?url=http://proquest.umi.com/pqdweb?did=725921521&sid=1&Fmt=2&clientId=2945&RQT=309&VName=PQD |
| Summary: | A new jackknife method is proposed for variance estimation of a heritability estimator in unbalanced random-effects models. This work is motivated by the need for analyzing a data set obtained in a project to study the quantitative genetics of fire ants. The new method uses the concept of weighted influence function by taking into account the unbalancedness of the data and allocates weights accordingly. Some theoretical properties of the new approach are investigated, and comparisons to existing methods that are developed for the balanced cases are discussed. Bootstrap methods are also explored for the distribution of the studentized statistic to construct sound confidence intervals. Monte Carlo simulation studies are conducted to examine the superior performance of the proposed method. Application of the proposed methodology to the fire ant data is demonstrated. The new method is also applied in the hypothesis testing problem with unbalanced data. Simulation studies are also conducted to compare it with the existing jackknife methods. |
|---|---|
| Item Description: | Vita. "Major Subject: Statistics". |
| Physical Description: | xi, 97 leaves : illustrations ; 28 cm. Issued also on microfiche from University Microfilm Inc. |
| Bibliography: | Includes bibliographical references (leaves 89-90). |