PEIV模型WTLS估计的Fisher-Score算法

Fisher-Score Algorithm of WTLS Estimation for PEIV Model

  • 摘要: 考虑系数矩阵含非随机元素和不同位置含相同随机元素的结构化特征,PEIV(partial errors-in-variables)模型较一般的EIV模型更为严格。现有PEIV模型加权整体最小二乘(weighted total least squares,WTLS)估计算法需多次迭代,影响计算效率。通过利用观测值误差和系数矩阵误差的统计性质构造非线性目标函数,并以此推导了新的PEIV模型WTLS估计的计算公式,同时设计了相应的Fisher-Score算法。算例分析结果表明,相比较而言,Fisher-Score算法迭代次数较少,计算效率得到大大提升。

     

    Abstract: Considering the non-random elements and the same random elements in different locations of coefficient matrix, the partial errors-in-variables(PEIV) model is stricter than general EIV model. However, the existing weighted total least squares (WTLS) algorithm for PEIV model requires more iteration so that the computation efficiency is reduced. Making using of statistical properties of observations errors and coefficient matrix error, this paper deduces a new computational formula of WTLS estimation for PEIV model based on Fisher-Score algorithm. The numerical results show that the Fisher-Score algorithm takes less iterations and the computation efficiency is enormously promoted.

     

/

返回文章
返回