Multivariate statistics : a vector space approach

By M. L. Eaton. 1985

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Sufficient Forecasting Using Factor Models (2015, background): “The linearity condition is satisﬁed when the distribution of f t is elliptically symmetric (Eaton, 1983), and it is also asymptotically justiﬁed when the dimension of f t is large (Hall and Li, 1993).”
Transformations which preserve exchangeability and application to permutation tests (2003, background): “31 of Eaton (1983), this implies that K 1⁄4 cE, where c is an isometric operator from R to R.”
Advanced Multivariate Statistics with Matrices (2005, methodology): “During a short time several monographs were published: Giri (1977), Srivastava & Khatri (1979), Mardia, Kent & Bibby (1979), Muirhead (1982), Takeuchi, Yanai & Mukherjee (1982), Eaton (1983), Farrell (1985) and Siotani, Hayakawa, & Fujikoshi (1985).”
On the mean and variance of the generalized inverse of a singular Wishart matrix (2011, background): “It is well known that this implies M11 = hpIp−1 for some real multiplier hp (see, for example, Eaton [5], Proposition 2.”
Analysis of multivariate repeated measures data with a Kronecker product structured covariance matrix (2001, background): “It is well known (for example, see Eaton, 1983, p. 305) that for any S ~ Wp(v, R ), E(S) 5 vR and D(S) 5 2v R R .”
All Invariant Moments of the Wishart Distribution (2004, background): “Muirhead (1982) and Eaton (1983). Haff (1982) gives the mean, covariance of U and also E(U) and E(U).”
The sparsity and bias of the Lasso selection in high-dimensional linear regression (2008, background): “Deﬁne For a ﬁxed P m , let Σ m be the m × m population covariance matrices of the rows of XP ′ m and 1 / Eaton (1983)].”
Spectral Radii of Large Non-Hermitian Random Matrices (2014, background): “It has a connection to the matrix F distribution in statistics literature; see, for instance, p. 331 from Eaton (2007).”
How to generate random matrices from the classical compact groups (2006, methodology): “It turns out that if the entries of Z ares i.i.d. standard complex normal random variables, i.e. if Z belongs to the Ginibre ensemble, then Q is distributed with Haar measure (see Eaton [3], p. 234, for a proof).”