方法对比

	稳健潜类别分析	鲁棒混合模型拟合
领域	统计学	统计学
方法族	Latent structure	Latent structure
起源年份	2000s	2000–2008
提出者	Building on Hennig (2004) and Vermunt & Magidson (2004)	Peel & McLachlan (t-mixture); Garcia-Escudero et al. (trimming framework)
类型	Robust latent variable / mixture model	Latent-class probabilistic clustering with outlier protection
开创性文献	Hennig, C. (2004). Breakdown points for maximum likelihood estimators of location-scale mixtures. Annals of Statistics, 32(4), 1313–1340.	Garcia-Escudero, L. A., Gordaliza, A., Matran, C. & Mayo-Iscar, A. (2008). A general trimming approach to robust cluster analysis. Annals of Statistics, 36(3), 1324–1345.
别名	robust LCA, outlier-resistant latent class analysis, trimmed-likelihood latent class analysis	robust mixture model, robust GMM, outlier-robust mixture model, trimmed mixture model
相关	6	5
摘要	Robust latent class analysis (robust LCA) extends the standard latent class model by incorporating outlier-resistant estimation techniques — such as trimmed likelihood, M-estimation, or downweighting — so that atypical response patterns do not distort the recovered class structure or class membership probabilities.	Robust mixture modeling fits finite mixture models — probabilistic clustering methods that assume data arise from a blend of underlying subpopulations — using component distributions or estimation strategies designed to be insensitive to outliers and heavy-tailed noise. The two dominant approaches replace Gaussian components with heavier-tailed distributions such as the multivariate t, or trim a fixed proportion of the most extreme observations before fitting.