多因子可加性实验模型参数估值及唯一性

    Estimate of Parameters and Its Uniqueness In a Model of Additional Experiment With Multifactors

    • 摘要: 提出了一个-般的n因子n-1级交互效应(n≥1)的可加性实验模型。如果所有的交互效应满足一定的约束条件,则用不完全求和算子,可证明母体均值、各级因子间交互效应及测量误差的估值是唯一的,从而证明了交互效应的计算是一组递推公式,易于编制计算机程序,可用于复杂系统的因子显著性检验及决策最优工程条件。

       

      Abstract: A model of additional experiment is suggested with n factors and up to n-1 interactions, when a certain set of constrained conditions is met, it can be proved by the incomplte-sum operator that the estimates of ensemble mean and all interactions and observation errors are defintiely unique. Meanwhile, it can be shown further that the calculation of all interreactions forms a recursive algorithm, which can be easily applied to a complex system for testing the significance of the factors and determining the optimal engineering conditions.

       

    /

    返回文章
    返回