手法を比較

	頑健な多変量分散分析	多変量分散分析 (MANOVA)
分野	統計学	統計学
系統	Hypothesis test	Hypothesis test
提唱年	1990s–2000s	1932
提唱者	Rand Wilcox; Lisa Lix and H. J. Keselman	Samuel Stanley Wilks (Wilks' Lambda, 1932); Roy, Hotelling, Pillai (mid-20th c.)
種類	Robust multivariate mean comparison	Parametric multivariate mean comparison
原典	Wilcox, R. R. (2012). Introduction to Robust Estimation and Hypothesis Testing (3rd ed.). Academic Press. ISBN: 978-0123869838	Tabachnick, B.G. & Fidell, L.S. (2013). Using Multivariate Statistics (6th ed.). Pearson. ISBN: 978-0205849574
別名	robust multivariate ANOVA, trimmed-mean MANOVA, outlier-resistant MANOVA, robust MANOVA	Multivariate ANOVA, Çok Değişkenli ANOVA (MANOVA)
関連	5	5
概要	Robust MANOVA is a multivariate analysis of variance procedure designed to remain valid when classical assumptions — multivariate normality and homogeneity of covariance matrices — are violated. It replaces raw means and standard covariance matrices with resistant estimates such as trimmed means and Winsorized covariances, yielding reliable Type I error control and power in the presence of outliers and skewed distributions across multiple dependent variables simultaneously.	MANOVA is a parametric hypothesis test that simultaneously compares group means across multiple continuous dependent variables, controlling the inflation of Type I error that would result from running separate ANOVAs. Key multivariate test statistics — Wilks' Lambda, Pillai's Trace, Hotelling-Lawley Trace, and Roy's Greatest Root — were developed between the 1930s and 1950s, with Wilks' Lambda formalised by Samuel Stanley Wilks in 1932.