| Summary: | 碩士 === 逢甲大學 === 應用數學所 === 99 === Breeders are usually interested in inferences on the ratio along with the
dominance and additive effects. This thesis considers the problem of finding
simultaneous confidence region for two regression coefficients and their ratio of
general linear models. We use the concept of generalized pivotal quantities to
construct the simultaneous confidence region including Plug-in (II), Bonferroni
correction and Generalized Variable approach. The proposed methods are compared
with the two traditional methods, Worst-case and Plug-in (I), based on the
multivariate-t distribution. A simulation study using the dominance ratio in crossing
experiments with plants is computed from an estimate of the dominance and additive
gene effects. Detailed statistical simulation studies are conducted to evaluate their
performance by the coverage rate. Furthermore, some practical examples are given to
illustrate the proposed procedures.
|
|---|
