Minimum Distance Estimation and Components of Goodness-Of-Fit Statistics
研究了最小距离估计与拟合优度统计量分量之间的关系，将M估计视为特例，通过Ψ函数与分量线性组合及高效得分的傅里叶近似解释，揭示了效率与稳健性权衡的本质在于高频分量系数的大小，并应用于复合和简单拟合优度问题。
Summary The relationship of minimum distance (MD) estimation to components of goodness-of-fit statistics is considered. M-estimation is viewed as a special case, with interesting interpretations in terms of the defining Ψ-function as related to linear combinations of the components and modified Fourier approximations to the efficient score. The essence of the competition between efficiency and robustness is the determination of the magnitudes of the coefficients of the high frequency components. Applications to the composite and simple goodness-of-fit problems are indicated.