eigenvectors of some large sample covariance matrix
play

Eigenvectors of some large sample covariance matrix ensembles - PowerPoint PPT Presentation

Feb 08, 2023 •286 likes •1.32k views

Eigenvectors of some large sample covariance matrix ensembles Random Matrix Workshop, Tlcom ParisTech Monday, October 11 th 2010 Olivier Ledoit Sandrine P ech e oledoit@iew.uzh.ch - sandrine.peche@ujf-grenoble.fr University of Z

  1. Generalization of MP67/S95 Equation � ( S N − zI ) − 1 � 1 m F N ( z ) = N Tr Eigenvectors of some large sample covariance matrix ensembles – p. 10/23

  2. Generalization of MP67/S95 Equation � ( S N − zI ) − 1 � 1 m F N ( z ) = N Tr � � 1 ( S N − zI ) − 1 g (Σ N ) Θ g N ( z ) = N Tr Eigenvectors of some large sample covariance matrix ensembles – p. 10/23

  3. Generalization of MP67/S95 Equation Eigenvectors of some large sample covariance matrix ensembles – p. 11/23

  4. Generalization of MP67/S95 Equation ∃ Θ g : a . s . Θ g ∀ z ∈ C + → Θ g ( z ) N ( z ) − Eigenvectors of some large sample covariance matrix ensembles – p. 11/23

  5. Generalization of MP67/S95 Equation ∃ Θ g : a . s . Θ g ∀ z ∈ C + → Θ g ( z ) N ( z ) − � + ∞ 1 Θ g ( z ) = g ( τ ) × τ [1 − c − czm F ( z )] − zdH ( τ ) −∞ Eigenvectors of some large sample covariance matrix ensembles – p. 11/23

  6. Generalization of MP67/S95 Equation ∃ Θ g : a . s . Θ g ∀ z ∈ C + → Θ g ( z ) N ( z ) − � + ∞ 1 Θ g ( z ) = g ( τ ) × τ [1 − c − czm F ( z )] − zdH ( τ ) −∞ Same integration kernel! Eigenvectors of some large sample covariance matrix ensembles – p. 11/23

  7. Extension to the Real Line Eigenvectors of some large sample covariance matrix ensembles – p. 12/23

  8. Extension to the Real Line N N � � 1 1 i v j | 2 × g ( τ j ) Θ g | u ∗ N ( z ) = N λ i − z i =1 j =1 Eigenvectors of some large sample covariance matrix ensembles – p. 12/23

  9. Extension to the Real Line N N � � 1 1 i v j | 2 × g ( τ j ) Θ g | u ∗ N ( z ) = N λ i − z i =1 j =1 N N � � 1 i v j | 2 × g ( τ j ) Ω g | u ∗ N ( λ ) = 1 [ λ i , + ∞ ) ( λ ) N i =1 j =1 Eigenvectors of some large sample covariance matrix ensembles – p. 12/23

  10. Extension to the Real Line N N � � 1 1 i v j | 2 × g ( τ j ) Θ g | u ∗ N ( z ) = N λ i − z i =1 j =1 N N � � 1 i v j | 2 × g ( τ j ) Ω g | u ∗ N ( λ ) = 1 [ λ i , + ∞ ) ( λ ) N i =1 j =1 � λ 1 Im [Θ g ( l + iη )] dl a . s . Ω g → Ω g ( λ ) = lim N ( λ ) π η → 0 + −∞ wherever Ω g is continuous Eigenvectors of some large sample covariance matrix ensembles – p. 12/23

  11. Sample Eigenvectors Eigenvectors of some large sample covariance matrix ensembles – p. 13/23

  12. Sample Eigenvectors Fix τ and take g = 1 ( −∞ ,τ ) Eigenvectors of some large sample covariance matrix ensembles – p. 13/23

  13. Sample Eigenvectors Fix τ and take g = 1 ( −∞ ,τ ) N N � � N ( λ ) = 1 i v j | 2 1 [ λ i , + ∞ ) ( λ ) × 1 [ τ j , + ∞ ) ( τ ) Ω g | u ∗ N i =1 j =1 � λ � τ clt → | t [1 − c − clm F ( l )] − l | 2 dH ( t ) dF ( l ) −∞ −∞ Eigenvectors of some large sample covariance matrix ensembles – p. 13/23

  14. Sample Eigenvectors Fix τ and take g = 1 ( −∞ ,τ ) N N � � N ( λ ) = 1 i v j | 2 1 [ λ i , + ∞ ) ( λ ) × 1 [ τ j , + ∞ ) ( τ ) Ω g | u ∗ N i =1 j =1 � λ � τ clt → | t [1 − c − clm F ( l )] − l | 2 dH ( t ) dF ( l ) −∞ −∞ cλ i τ j i v j | 2 ≈ N | u ∗ m F ( λ i )] − λ i | 2 | τ j [1 − c − cλ i ˘ Eigenvectors of some large sample covariance matrix ensembles – p. 13/23

  15. Eigenvalues with Multiplicity Eigenvectors of some large sample covariance matrix ensembles – p. 14/23

  16. Eigenvalues with Multiplicity Σ N has K distinct eigenvalues t 1 , . . . , t K with multiplicities n 1 , . . . , n K Eigenvectors of some large sample covariance matrix ensembles – p. 14/23

  17. Eigenvalues with Multiplicity Σ N has K distinct eigenvalues t 1 , . . . , t K with multiplicities n 1 , . . . , n K P k = projection onto k th eigenspace Eigenvectors of some large sample covariance matrix ensembles – p. 14/23

  18. Eigenvalues with Multiplicity Σ N has K distinct eigenvalues t 1 , . . . , t K with multiplicities n 1 , . . . , n K P k = projection onto k th eigenspace n k cλ i t k | P k u i | 2 ≈ m F ( λ i )] − λ i | 2 N | t k [1 − c − cλ i ˘ Eigenvectors of some large sample covariance matrix ensembles – p. 14/23

  19. Estimating the Covariance Matrix (1) Eigenvectors of some large sample covariance matrix ensembles – p. 15/23

  20. Estimating the Covariance Matrix (1) � Tr ( AA ∗ ) Frobenius norm: � A � = Eigenvectors of some large sample covariance matrix ensembles – p. 15/23

  21. Estimating the Covariance Matrix (1) � Tr ( AA ∗ ) Frobenius norm: � A � = U N : matrix of eigenvectors of S N Eigenvectors of some large sample covariance matrix ensembles – p. 15/23

  22. Estimating the Covariance Matrix (1) � Tr ( AA ∗ ) Frobenius norm: � A � = U N : matrix of eigenvectors of S N Find matrix closest to Σ N among those that have eigenvectors U N Eigenvectors of some large sample covariance matrix ensembles – p. 15/23

  23. Estimating the Covariance Matrix (1) � Tr ( AA ∗ ) Frobenius norm: � A � = U N : matrix of eigenvectors of S N Find matrix closest to Σ N among those that have eigenvectors U N � U N D N U ∗ min N − Σ N � D N diagonal Eigenvectors of some large sample covariance matrix ensembles – p. 15/23

  24. Estimating the Covariance Matrix (1) � Tr ( AA ∗ ) Frobenius norm: � A � = U N : matrix of eigenvectors of S N Find matrix closest to Σ N among those that have eigenvectors U N � U N D N U ∗ min N − Σ N � D N diagonal Solution: D N = Diag ( � � d 1 , . . . , � � d i = u ∗ d N ) where i Σ N u i Eigenvectors of some large sample covariance matrix ensembles – p. 15/23

  25. Estimating the Covariance Matrix (2) Eigenvectors of some large sample covariance matrix ensembles – p. 16/23

  26. Estimating the Covariance Matrix (2) Take g ( τ ) = τ Eigenvectors of some large sample covariance matrix ensembles – p. 16/23

  27. Estimating the Covariance Matrix (2) Take g ( τ ) = τ N � N ( λ ) = 1 Ω g u ∗ i Σ N u i 1 [ λ i , + ∞ ) ( λ ) N i =1 � λ l → | 1 − c − clm F ( l ) | 2 dF ( l ) −∞ Eigenvectors of some large sample covariance matrix ensembles – p. 16/23

  28. Estimating the Covariance Matrix (2) Take g ( τ ) = τ N � N ( λ ) = 1 Ω g u ∗ i Σ N u i 1 [ λ i , + ∞ ) ( λ ) N i =1 � λ l → | 1 − c − clm F ( l ) | 2 dF ( l ) −∞ λ i u ∗ i Σ N u i ≈ m F ( λ i ) | 2 | 1 − c − cλ i ˘ Eigenvectors of some large sample covariance matrix ensembles – p. 16/23

  29. Oracle Estimator Eigenvectors of some large sample covariance matrix ensembles – p. 17/23

  30. Oracle Estimator Keep same eigenvectors as those of S n , Eigenvectors of some large sample covariance matrix ensembles – p. 17/23

  31. Oracle Estimator Keep same eigenvectors as those of S n , divide i th sample eigenvalue by | 1 − c − cλ i ˘ m F ( λ i ) | 2 Eigenvectors of some large sample covariance matrix ensembles – p. 17/23

  32. Oracle Estimator Keep same eigenvectors as those of S n , divide i th sample eigenvalue by | 1 − c − cλ i ˘ m F ( λ i ) | 2 → oracle estimator � − S N Eigenvectors of some large sample covariance matrix ensembles – p. 17/23

  33. Oracle Estimator Keep same eigenvectors as those of S n , divide i th sample eigenvalue by | 1 − c − cλ i ˘ m F ( λ i ) | 2 → oracle estimator � − S N Percentage Relative Improvement in Average Loss:   � � 2 � � �� S N − U N � D N U ∗ E � N   PRIAL = 100 ×  1 − � �  2 � � � S N − U N � D N U ∗ E � N Eigenvectors of some large sample covariance matrix ensembles – p. 17/23

  34. Monte-Carlo Simulations Eigenvectors of some large sample covariance matrix ensembles – p. 18/23

  35. Monte-Carlo Simulations 10,000 simulations Eigenvectors of some large sample covariance matrix ensembles – p. 18/23

  36. Monte-Carlo Simulations 10,000 simulations c=1/2 Eigenvectors of some large sample covariance matrix ensembles – p. 18/23

  37. Monte-Carlo Simulations 10,000 simulations c=1/2 Population eigenvalues: 20% equal to 1 40% equal to 3 40% equal to 10 Eigenvectors of some large sample covariance matrix ensembles – p. 18/23

  38. Monte-Carlo Simulations 10,000 simulations c=1/2 Population eigenvalues: 20% equal to 1 40% equal to 3 40% equal to 10 Compare with Ledoit-Wolf (2004) linear shrinkage estimator Eigenvectors of some large sample covariance matrix ensembles – p. 18/23

  39. Simulation Results Eigenvectors of some large sample covariance matrix ensembles – p. 19/23

  40. Simulation Results γ=2 100% Relative Improvement in Average Loss 90% 80% 70% 60% Optimal Nonlinear Shrinkage Optimal Linear Shrinkage 50% 10 20 40 60 80 100 120 140 160 180 200 Sample Size Eigenvectors of some large sample covariance matrix ensembles – p. 19/23

  41. Inverse of the Covariance Matrix (1) Eigenvectors of some large sample covariance matrix ensembles – p. 20/23

  42. Inverse of the Covariance Matrix (1) Find matrix closest to Σ − 1 N among those that have eigenvectors U N Eigenvectors of some large sample covariance matrix ensembles – p. 20/23

  43. Inverse of the Covariance Matrix (1) Find matrix closest to Σ − 1 N among those that have eigenvectors U N � U N ∆ N U ∗ N − Σ − 1 min N � ∆ N diagonal Eigenvectors of some large sample covariance matrix ensembles – p. 20/23

  44. Inverse of the Covariance Matrix (1) Find matrix closest to Σ − 1 N among those that have eigenvectors U N � U N ∆ N U ∗ N − Σ − 1 min N � ∆ N diagonal Solution: ∆ N = Diag ( � � δ 1 , . . . , � � δ i = u ∗ i Σ − 1 δ N ) where N u i Eigenvectors of some large sample covariance matrix ensembles – p. 20/23

  45. Inverse of the Covariance Matrix (1) Find matrix closest to Σ − 1 N among those that have eigenvectors U N � U N ∆ N U ∗ N − Σ − 1 min N � ∆ N diagonal Solution: ∆ N = Diag ( � � δ 1 , . . . , � � δ i = u ∗ i Σ − 1 δ N ) where N u i i Σ N u i ) − 1 u ∗ i Σ − 1 N u i ≥ ( u ∗ Eigenvectors of some large sample covariance matrix ensembles – p. 20/23

  46. Inverse of the Covariance Matrix (2) Eigenvectors of some large sample covariance matrix ensembles – p. 21/23

  47. Inverse of the Covariance Matrix (2) Take g ( τ ) = 1 τ Eigenvectors of some large sample covariance matrix ensembles – p. 21/23

  48. Inverse of the Covariance Matrix (2) Take g ( τ ) = 1 τ N � N ( λ ) = 1 Ω g u ∗ i Σ − 1 N u i 1 [ λ i , + ∞ ) ( λ ) N i =1 � λ 1 − c − 2 cl Re [ m F ( l )] → dF ( l ) l −∞ Eigenvectors of some large sample covariance matrix ensembles – p. 21/23

  49. Inverse of the Covariance Matrix (2) Take g ( τ ) = 1 τ N � N ( λ ) = 1 Ω g u ∗ i Σ − 1 N u i 1 [ λ i , + ∞ ) ( λ ) N i =1 � λ 1 − c − 2 cl Re [ m F ( l )] → dF ( l ) l −∞ N u i ≈ 1 − c − 2 cλ i Re [ ˘ m F ( λ i )] u ∗ i Σ − 1 λ i Eigenvectors of some large sample covariance matrix ensembles – p. 21/23

  50. Conclusion Eigenvectors of some large sample covariance matrix ensembles – p. 22/23

  51. Conclusion Generalization of the Marˇ cenko-Pastur (1967)/Silverstein (1995) Equation Eigenvectors of some large sample covariance matrix ensembles – p. 22/23

  52. Conclusion Generalization of the Marˇ cenko-Pastur (1967)/Silverstein (1995) Equation Gives location of sample eigenvectors relative to: Eigenvectors of some large sample covariance matrix ensembles – p. 22/23

  53. Conclusion Generalization of the Marˇ cenko-Pastur (1967)/Silverstein (1995) Equation Gives location of sample eigenvectors relative to: Population eigenvectors Eigenvectors of some large sample covariance matrix ensembles – p. 22/23

  54. Conclusion Generalization of the Marˇ cenko-Pastur (1967)/Silverstein (1995) Equation Gives location of sample eigenvectors relative to: Population eigenvectors Population covariance matrix as a whole Eigenvectors of some large sample covariance matrix ensembles – p. 22/23

  55. Conclusion Generalization of the Marˇ cenko-Pastur (1967)/Silverstein (1995) Equation Gives location of sample eigenvectors relative to: Population eigenvectors Population covariance matrix as a whole Inverse of population covariance matrix Eigenvectors of some large sample covariance matrix ensembles – p. 22/23

  56. Conclusion Generalization of the Marˇ cenko-Pastur (1967)/Silverstein (1995) Equation Gives location of sample eigenvectors relative to: Population eigenvectors Population covariance matrix as a whole Inverse of population covariance matrix We do for sample eigenvectors what MP67/S95 did for sample eigenvalues Eigenvectors of some large sample covariance matrix ensembles – p. 22/23

  57. Directions for Future Research Eigenvectors of some large sample covariance matrix ensembles – p. 23/23

  58. Directions for Future Research 1. Construct bona fide nonlinear shrinkage estimator of the covariance matrix Eigenvectors of some large sample covariance matrix ensembles – p. 23/23

Recommend

Matrix Algebra of Sample Statistics James H. Steiger Department of Psychology and Human

Matrix Algebra of Sample Statistics James H. Steiger Department of Psychology and Human

Matrix Algebra of Some Sample Statistics Variance of a Linear Combination Variance-Covariance Matrix of Several Linear Combinations Covariance Matrix of Two Sets of Linear Combinations Matrix Algebra of Sample Statistics James H. Steiger

338 views • 29 slides
Least Squares Estimation- Large-Sample Properties Ping Yu School of Economics and Finance The

Least Squares Estimation- Large-Sample Properties Ping Yu School of Economics and Finance The

Least Squares Estimation- Large-Sample Properties Ping Yu School of Economics and Finance The University of Hong Kong Ping Yu (HKU) Large-Sample 1 / 63 Asymptotics for the LSE 1 Covariance Matrix Estimators 2 Functions of Parameters 3

881 views • 64 slides
TracyWidom limit for sample covariance matrices Kevin Schnelli KTH Royal Institute of

TracyWidom limit for sample covariance matrices Kevin Schnelli KTH Royal Institute of

TracyWidom limit for sample covariance matrices Kevin Schnelli KTH Royal Institute of Technology Joint work with Jong Yun Hwang (KAIST) and Ji Oon Lee (KAIST) May 10, 2019 Sample covariance matrix Consider a (mean-zero) multivariate

589 views • 28 slides
Eigenvalues, Eigenvectors, Matrix Factoring, and Principal Components The eigenvalues and

Eigenvalues, Eigenvectors, Matrix Factoring, and Principal Components The eigenvalues and

Eigenvalues, Eigenvectors, Matrix Factoring, and Principal Components The eigenvalues and eigenvectors of a square matrix play a key role in some important operations in statistics. In particular, they are intimately connected with the

377 views • 25 slides
Covariance Matrix Adaptation Covariance Matrix Adaptation Evolution Strategies Recalling New

Covariance Matrix Adaptation Covariance Matrix Adaptation Evolution Strategies Recalling New

Covariance Matrix Adaptation Covariance Matrix Adaptation Evolution Strategies Recalling New search points are sampled normally distributed x i m + N i ( 0 , C ) for i = 1 , . . . , where x i , m 2 R n , 2 R + , as perturbations of

1.07k views • 52 slides
Eigenvalues and Eigenvectors Let A R n n be a matrix. If R and v R n , v = 0,

Eigenvalues and Eigenvectors Let A R n n be a matrix. If R and v R n , v = 0,

Eigenvalues and Eigenvectors Let A R n n be a matrix. If R and v R n , v = 0, with Av = v , then we call 1. an eigenvalue of A , 2. v an eigenvector of A , 3. and ( , v ) an eigenpair of A Eigen, adjective: own,

730 views • 51 slides
Random Matrix Improved Covariance Estimation for a Large Class of Metrics Malik TIOMOKO , Florent

Random Matrix Improved Covariance Estimation for a Large Class of Metrics Malik TIOMOKO , Florent

Random Matrix Improved Covariance Estimation for a Large Class of Metrics Malik TIOMOKO , Florent BOUCHARD, Guillaume GINOLHAC and Romain COUILLET GSTATS IDEX DataScience Chair, GIPSA-lab, University GrenobleAlpes, France. Laboratoire des

409 views • 14 slides
Matrix Calculations: Eigenvalues and Eigenvectors H. Geuvers (and A. Kissinger) Institute for

Matrix Calculations: Eigenvalues and Eigenvectors H. Geuvers (and A. Kissinger) Institute for

Eigenvalues and Eigenvectors Radboud University Nijmegen Applications of Eigenvalues and Eigenvectors Matrix Calculations: Eigenvalues and Eigenvectors H. Geuvers (and A. Kissinger) Institute for Computing and Information Sciences Radboud

923 views • 35 slides
Visualization 1 Applied Multivariate Statistics Spring 2012 Goals Covariance, Correlation

Visualization 1 Applied Multivariate Statistics Spring 2012 Goals Covariance, Correlation

Visualization 1 Applied Multivariate Statistics Spring 2012 Goals Covariance, Correlation (true / sample version) Test for zero correlation: Fishers z -Transformation Scatterplot / Scatterplotmatrix Covariance matrix /

455 views • 21 slides
Analytical Nonlinear Shrinkage of Large-Dimensional Covariance Matrices Olivier Ledoit 1 and

Analytical Nonlinear Shrinkage of Large-Dimensional Covariance Matrices Olivier Ledoit 1 and

Introduction Finite Samples Random Matrix Theory Kernel Estimation Monte Carlo Application Conclusion Analytical Nonlinear Shrinkage of Large-Dimensional Covariance Matrices Olivier Ledoit 1 and Michael Wolf 1 1 Department of Economics

1.99k views • 164 slides
Eigenvalues and Eigenvectors Eigenvalue problem Let ! be an "" matrix: $ & is

Eigenvalues and Eigenvectors Eigenvalue problem Let ! be an "" matrix: $ & is

Eigenvalues and Eigenvectors Eigenvalue problem Let ! be an "" matrix: $ & is an eigenvector of ! if there exists a scalar ' such that ! $ = ' $ I l eigenvectors where ' is called an eigenvalue . eigenvalues eigeipairs If $ is

723 views • 13 slides
Week 1: Introduc/on Precision and covariance matrix 2 1.2C

Week 1: Introduc/on Precision and covariance matrix 2 1.2C

1 Week 1: Introduc/on Precision and covariance matrix 2 1.2C Precision and covariance matrix 3 Histogram of random errors high precision Density 0 low precision 0 Random

308 views • 13 slides
Eigenvalues and Eigenvectors Suppose A is an n n symmetric matrix with real entries. The

Eigenvalues and Eigenvectors Suppose A is an n n symmetric matrix with real entries. The

Eigenvalues and Eigenvectors Suppose A is an n n symmetric matrix with real entries. The function from R n to R defined by x x t Ax is called a quadratic form. We can maximize x T Ax subject to x T x = || x || 2 = 1 by

742 views • 33 slides
Implementation of Covariance Matrix on ReconstructedParticle C. Calancha ILD Analysis &

Implementation of Covariance Matrix on ReconstructedParticle C. Calancha ILD Analysis &

Implementation of Covariance Matrix on ReconstructedParticle C. Calancha ILD Analysis & Software Meeting April 16, 2014 C. Calancha (KEK) Covariance Matrix April 16, 2014 1/11 Motivation ReconstructedParticle.getCovMatrix is not

443 views • 11 slides
Multivariate Gaussian Mean vector: Covariance matrix: 2 1 Conditioning a Gaussian Joint

Multivariate Gaussian Mean vector: Covariance matrix: 2 1 Conditioning a Gaussian Joint

Readings: K&F: 6.1, 6.2, 6.3, 14.1, 14.2, 14.3, 14.4, Kalman Filters Gaussian MNs Graphical Models 10708 Carlos Guestrin Carnegie Mellon University December 1 st , 2008 1 Multivariate Gaussian Mean vector: Covariance matrix: 2 1

257 views • 11 slides
= an orthogonal matrix T X X l l Cov( , ) i k ij kj j = 1 = X F l Cov( , )

= an orthogonal matrix T X X l l Cov( , ) i k ij kj j = 1 = X F l Cov( , )

Factor analysis (cf. section 9.3) In matrix notation the factor model takes the form: An observable random vector X of dimension p has mean X LF = + vector and covariance matrix L Here is the p x m matrix

98 views • 6 slides
Spiked Eigenvalues of High Dimensional Separable Sample Covariance Matrices Guangming Pan,

Spiked Eigenvalues of High Dimensional Separable Sample Covariance Matrices Guangming Pan,

Spiked Eigenvalues of High Dimensional Separable Sample Covariance Matrices Guangming Pan, Nanyang Technological University, Singapore November 19, 2019 Guangming Pan, (USTC) Spiked Eigenvalues of High Dimensional Separable Sample Covariance

930 views • 76 slides
Responses on Sample Covariance G+SSC+cNG Cov(k1,k2) Alexandre Barreira MPA P(k) R(k) T (

Responses on Sample Covariance G+SSC+cNG Cov(k1,k2) Alexandre Barreira MPA P(k) R(k) T (

Responses on Sample Covariance G+SSC+cNG Cov(k1,k2) Alexandre Barreira MPA P(k) R(k) T ( k 1 , - k 1 with Elisabeth Krause & , k 2 , - k 2 ) Fabian Schmidt arXiv:1703.09212 arXiv:1705.01092 arXiv:1711.07467

617 views • 59 slides
Eigenvalues and Eigenvectors Eigenvalue problem Let be an matrix: is

Eigenvalues and Eigenvectors Eigenvalue problem Let be an matrix: is

Eigenvalues and Eigenvectors Eigenvalue problem Let be an matrix: is an eigenvector of if there exists a scalar such that = where is called an eigenvalue . If is an eigenvector,

338 views • 13 slides
Overview Last week introduced the important Diagonalisation Theorem: An n n matrix A is

Overview Last week introduced the important Diagonalisation Theorem: An n n matrix A is

Overview Last week introduced the important Diagonalisation Theorem: An n n matrix A is diagonalisable if and only if there is a basis for R n consisting of eigenvectors of A. This week well continue our study of eigenvectors and

860 views • 69 slides
Latent Semantic Analysis (Tutorial) Alex Thomo 1 Eigenvalues and Eigenvectors Let A be an n

Latent Semantic Analysis (Tutorial) Alex Thomo 1 Eigenvalues and Eigenvectors Let A be an n

Latent Semantic Analysis (Tutorial) Alex Thomo 1 Eigenvalues and Eigenvectors Let A be an n n matrix with elements being real numbers. If x is an n -dimensional vector, then the matrix-vector product A x is well-defined, and the result is

221 views • 7 slides
Approximating the covariance matrix with heavy tailed columns and RIP. Alexander Litvak

Approximating the covariance matrix with heavy tailed columns and RIP. Alexander Litvak

Approximating the covariance matrix with heavy tailed columns and RIP. Alexander Litvak University of Alberta based on a joint work with O. Gudon, A. Pajor and N. Tomczak-Jaegermann (the paper On the interval ... available at:

853 views • 71 slides
Log Covariance Matrix Estimation Xinwei Deng Department of Statistics University of

Log Covariance Matrix Estimation Xinwei Deng Department of Statistics University of

Log Covariance Matrix Estimation Xinwei Deng Department of Statistics University of Wisconsin-Madison Joint work with Kam-Wah Tsui (Univ. of Wisconsin-Madsion) 1 Outline Background and Motivation The Proposed Log-ME Method

518 views • 27 slides
1st, choose a covariance model; 2nd, aprroximate the precision matrix Q ; 3rd, draw approximate

1st, choose a covariance model; 2nd, aprroximate the precision matrix Q ; 3rd, draw approximate

1st, choose a covariance model; 2nd, aprroximate the precision matrix Q ; 3rd, draw approximate inference. Henrique Laureano http://leg.ufpr.br/~henrique December 16, 2019 spde2smoothing Where? Journal of Agricultural, Biological, and

321 views • 12 slides

More recommend