摘要
经纬杂交试验主要涉及地方品种。这些品种具有两个显著特点:一是品种之内纯合程度较差,二是品种之间纯合程度不同。所以对于不同杂交组合而言,相应于观测值的误差效应不仅很大而且服从不同的分布;此时若再利用正规最小二乘进行分析,必然不能得到正确结果。本文根据客观特点,依据新的模型和最优线性无偏估计原理,对于这种试验提出了一种新的分析方法——一般最小二乘分析。这种方法只需首先估计各个组合的方差,然后用它对重复数和观测值进行校正,再按正规最小二乘分析步骤分析即可,所以并不复杂。具体使用有一实例说明。
Two-way network crosses are often related to indigenous breeds which have two notabl characteristics: 1. individuals within-breed are different, 2. the differences between-breeds are different too. Because of this, regular Least-square analysis method can not be used in this type of test. This paper proposed a new method, which was based on the new model and the best liner unbiased estimation theory. An example was given to show the usage.
