方法对比

	稳健多维尺度分析 (Robust MDS)	稳健探索性因子分析
领域	统计学	心理测量学
方法族	Latent structure	Latent structure
起源年份	2002 (robust extension); 1952 (classical MDS)	2000–2003
提出者	Hubert, Arabie, and Meulman (robust extensions); classical MDS by Torgerson (1952)	Pison, Rousseeuw, Filzmoser, and Croux; Yuan and Bentler (parallel streams)
类型	Dimensionality reduction / proximity scaling	Latent variable / dimension reduction (robust)
开创性文献	Hubert, L., Arabie, P. & Meulman, J. (2002). Linear unidimensional scaling in the L2-norm: Basic optimization methods using SMACOF. Journal of Classification, 19(2), 303–327.	Yuan, K.-H., & Bentler, P. M. (2000). Robust mean and covariance structure analysis through iteratively reweighted least squares. Psychometrika, 65(1), 43–58.
别名	Robust MDS, outlier-resistant MDS, robust proximity scaling	robust EFA, robust factor analysis, outlier-resistant factor analysis, EFA with robust estimation
相关	4	4
摘要	Robust multidimensional scaling recovers a low-dimensional spatial map from a matrix of pairwise dissimilarities while resisting distortion caused by outlying or erroneous proximity values. By replacing squared-error loss with a robust loss function or down-weighting suspect pairs, it produces a configuration that faithfully represents the bulk of the data even when some distances are grossly atypical.	Robust exploratory factor analysis discovers the latent factor structure of a set of items using estimation methods that are resistant to outliers and violations of multivariate normality. It applies the same measurement model as standard EFA but replaces classical covariance estimation with robust counterparts — such as minimum covariance determinant or iteratively reweighted least squares — so that a small fraction of atypical cases cannot distort the recovered factor loadings.