Pith. sign in

REVIEW 2 major objections 4 minor 161 references

Adaptable Regularized CCA Tests for Independence of High-Dimensional Random Vectors

T0 review · 2 major / 4 minor · reviewed 2026-07-14 · grok-4.5

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Ridge-and-PC regularized CCA tests give valid high-dimensional independence tests with normal or Tracy–Widom limits according to the reduced dimension.

desk verdict Solid, usable high-dim CCA independence tests that finally cover both p1 and p2 ≳ n, with clean RMT null limits and a principled ridge selector; the PC-capture modeling price is stated honestly and does not break the claims. read the letter →

arxiv 2607.10500 v1 pith:BYZWF2Z7 submitted 2026-07-11 stat.ME

Haoran Li This is my paper
classification stat.ME MSC 62H2062H1560B20
keywords canonicalcorrelationanalysishigh-dimensionalindependencetestridgeregularizationprincipalcomponentdimensionreductionTracy–Widomlawregularizedlikelihoodratiolargest-rootrandommatrixtheory
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Classical CCA independence tests break when dimensions of X and Y are comparable to or larger than sample size, because the relevant covariance matrices become singular or nearly singular. This paper stabilizes those tests by first projecting onto the leading sample principal components of one vector and then applying ridge regularization to the residual covariance of the other. When the number of retained components stays fixed, a regularized trace statistic is asymptotically standard normal after simple plug-in centering and scaling; when that number grows proportionally with sample size, the largest eigenvalue converges to the Tracy–Widom law after analogous normalization. Both procedures remain well-defined for arbitrary dimensions, are rotation-invariant, and are consistent against alternatives whose dependence is captured by a modest number of leading principal components. A Bayesian–minimax rule supplies a data-driven choice of the ridge parameter, and simulations show reliable size and competitive power across identity, AR, and spiked covariance structures.

What carries the argument

The ridge-regularized, rank-k F-matrix F_{kλ}=W_{k1}(W_{k2}+λI)^{-1}, obtained by projecting onto the leading k sample principal components of Y and ridge-stabilizing the residual covariance of X; its eigenvalues become the classical CCA roots when λ=0 and k equals the full dimension of Y.

What would settle it

Generate high-dimensional data under a dense alternative whose cross-covariance has many comparable singular values with no isolated spikes; if the proposed tests then lose power relative to unregularized or nonparametric competitors while still controlling size under the null, the central claim about practical utility fails.

Watch

Extended reading notes

Core claim

Under the high-dimensional regime and a model in which any dependence between X and Y is linear through the leading principal components of Y, the ridge-and-PC regularized F-matrix yields two asymptotically pivotal statistics: a fixed-k trace that is standard normal after data-driven centering and scaling, and a proportional-k largest root that is Tracy–Widom after analogous normalization; both tests are consistent against alternatives satisfying a mild signal-strength condition.

Load-bearing premise

Any dependence between the two vectors must be carried by a modest number of leading principal components of one of them; if the dependence lives in many weak bulk directions, the power guarantees and the recommended choice of reduced dimension no longer hold.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 4 minor

Summary. The paper develops ridge-regularized, PC-reduced CCA tests for independence of two high-dimensional vectors X and Y under the regime p1 ≍ n and p2 ≍ n^ζ (ζ ≥ 1). Dependence, if present, is modeled as linear through the leading m principal components of Y (Model (1.1)). After projecting onto the leading k sample PCs of Y and ridge-regularizing the residual Gram matrix, the authors form a regularized F-matrix F_{kλ}. For fixed k they obtain a normal limit for the trace statistic T(k,λ) after data-driven centering and scaling (Theorems 3.4–3.5); for k/n → γ ∈ (0,α) they obtain a Tracy–Widom limit for the largest root (Theorem 3.7). Consistency under alternatives that satisfy a residual-signal vanishing condition (4.1) and a mild signal-strength lower bound is proved (Theorems 4.1–4.2). A Bayesian–minimax selector for λ on a finite grid is shown to be consistent (Lemma 5.3), and practical rules for designating X versus Y and for choosing k are supplied. Finite-sample size and power are examined against two existing CCA procedures.

Significance. The work fills a genuine gap: classical and existing high-dimensional CCA tests become undefined or unstable once both p1 and p2 may exceed n. By combining ridge regularization with PC dimension reduction the authors obtain well-defined procedures whose null limits rest on established RMT local laws and Green-function comparison (Knowles–Yin, Han et al., Li 2025). The dual asymptotic regimes (fixed-k normal / diverging-k Tracy–Widom) and the accompanying consistent estimators of the centering and scaling constants are technically solid and immediately usable. The data-driven λ selector is protected by a finite-grid consistency argument, and the simulation design covers identity, AR(1) and polynomial spectra as well as low-rank and exponentially decaying alternatives. If the modeling assumption that dependence lives in the leading PCs of Y is accepted, the paper supplies a practical, theoretically justified toolkit for a problem that arises routinely in genomics, finance and network analysis.

major comments (2)
  1. The power analysis and the practical recommendation for k rest on Condition (4.1) (the residual signal after projection onto the leading k sample PCs of Y vanishes in operator norm). Theorems 4.1–4.2 and the rank-one characterizations (4.4)–(4.5) are correctly stated as conditional on this assumption, yet the manuscript never quantifies how large k must be relative to the unknown intrinsic rank m, nor does it supply a diagnostic that would warn a user when (4.1) fails. A short numerical illustration (or a theoretical bound) showing the degradation of power when dependence lives in the bulk would make the scope of the consistency claims clearer and would strengthen the practical guidance in Section 6.
  2. Section 5 selects λ by maximizing a Bayes–minimax proxy SNR(λ,q) that is derived under a rank-one alternative. While Lemma 5.3 guarantees that the finite-grid selector does not disturb the null limits, the paper provides no evidence that the same selector remains near-optimal under the multi-spike or exponentially decaying alternatives used in the simulations. A brief comparison of power under the data-driven λ versus an oracle λ chosen to maximize empirical power under those alternatives would confirm that the Bayesian–minimax criterion is not overly specialized.
minor comments (4)
  1. Table 7.1 shows mild size inflation for the trace statistic once k/p2 ≥ 0.2; the text already recommends k ≲ 20, but an explicit numerical cutoff (or a rule based on the estimated aspect ratio q) would help practitioners.
  2. The notation for the deterministic equivalent D(z) and the Stieltjes transform φ(z) is introduced in Section 3.1 and then reused with suppressed arguments; a short “notation reminder” box or a consistent subscripting convention would improve readability.
  3. Figures 7.1–7.3 and B.1–B.3 are dense; adding a common legend and slightly thicker lines for the proposed procedures would make the power comparisons easier to parse.
  4. A few typographical inconsistencies appear (e.g., “Y ang” versus “Yang”, occasional missing spaces around mathematical operators). A careful copy-edit pass is recommended.

Circularity Check

1 steps flagged · score 1.0 of 10

No significant circularity: asymptotics rest on external RMT plus explicit adaptations of the author's prior ridge-regularized results; data-driven λ is protected by a finite-grid consistency lemma rather than being a fitted quantity re-labeled as prediction.

  1. self citation load bearing [Remark 3.4 and Appendix A (estimation of Θ1,Θ2)]
    "When Y is treated as deterministic, Theorem 3.7 specializes to the main theorem of Li (2025). … The procedure is adapted from Li (2025), with several modifications tailored to the present framework. … The results are adapted from Theorem 3.2, Lemma 3.3, and Lemma 3.4 of Li (2025) whose proofs are therefore omitted."

    The Tracy–Widom centering/scaling constants and their consistent estimators for the largest-root statistic are obtained by specializing and adapting the author's own prior ridge-regularized result rather than being re-derived from first principles inside the present paper. The step is load-bearing for the practical implementation of ℓ_max(k,λ) but is not circular in the strong sense: the prior work is an independent RMT theorem under a related model, the present paper supplies the additional arguments needed for random Y and the PC projection, and the null limit itself is not defined in terms of the estimator.

full rationale

The derivation chain for the strongest claims (Theorems 3.5 and 3.7 under the null, and consistency Theorems 4.1–4.2 under the stated signal condition) proceeds from classical CCA F-matrix representations, ridge-regularized resolvents, and standard RMT local laws / edge universality (Knowles–Yin, Lee–Schnelli, Tracy–Widom, etc.). Self-citations to Li (2025) and Li et al. (2020a,b) supply technical lemmas on ridge-regularized largest roots, linear spectral statistics, and estimators of the Marčenko–Pastur functionals H_j; these are used as building blocks and are re-proved or adapted in the appendix under the present CCA projection. They do not define the target statistics in terms of themselves, nor do they import an unverified uniqueness theorem that forces the present conclusions. The Bayesian–minimax selector for λ maximizes an estimated risk whose consistency (Lemma 5.3) is proved under the same null/alternative conditions already used for the tests; once the finite candidate set L is fixed, the selected λ is asymptotically non-random and the null limits remain valid. No fitted parameter is later called a prediction, no ansatz is smuggled via citation, and no known empirical pattern is merely renamed. The modeling assumption (1.1)+(4.1) is an explicit scope restriction, not a circular step. Hence the paper is essentially self-contained against external benchmarks; the single minor self-citation layer raises the score only to 1.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The central claims rest on a structural model that funnels all dependence through leading PCs of one vector, on standard high-dimensional RMT regularity conditions, and on two free tuning parameters (k and λ) that are selected by data-driven but still heuristic rules. No new physical entities are postulated; the free parameters are the usual regularization and rank choices of the method.

free parameters (2)
  • regularization parameter λ
    Chosen by a Bayesian-minimax criterion over a discrete grid of 10 points between (0.1q)∧1 times average residual variance and twice that value; the grid and the prior class (θ=1) are user-specified.
  • reduced dimension k
    Selected by a sequential stability rule on the residual trace (stop when relative change ≤5 %) inside the window [min(0.01n,p2), min(0.5n,p2)]; the thresholds 0.05, 0.01, 0.5 are fixed by the author.
assumptions (4)
  • domain assumption Model (1.1): dependence between X and Y is exactly linear through the first m eigenvectors of Σy (m may grow but is captured by the chosen k).
    Stated in Section 1; all subsequent power analysis and the practical designation of which vector is Y rest on it.
  • domain assumption Conditions C1–C5 (high-dimensional regime, all moments of Zx finite, bounded spectrum of Σ0, sufficient rank of Sy, edge regularity of Σ0 near its largest eigenvalue).
    Section 3; required for the local laws and Tracy–Widom limits.
  • standard math Standard RMT local laws and Green-function comparison for sample covariance and F-type matrices (Knowles–Yin 2017, Han et al. 2018, Li 2025).
    Invoked throughout Appendix C; the paper adapts rather than re-proves them.
  • ad hoc to paper Condition (4.1): the residual signal after projection onto the leading k sample PCs of Y vanishes in operator norm.
    Section 4; needed for consistency of both tests and for consistency of the normalization estimators under the alternative.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Adaptable Regularized CCA Tests for Independence of High-Dimensional Random Vectors." pith.science (2026). https://pith.science/paper/BYZWF2Z7

@misc{pith2026260710500,
  author       = {Pith},
  title        = {Pith review of: Adaptable Regularized CCA Tests for Independence of High-Dimensional Random Vectors},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BYZWF2Z7}},
  note         = {Machine review of arXiv:2607.10500}
}
read the original abstract

We propose an adaptable testing procedure for independence between two high-dimensional random vectors. The method incorporates ridge regularization and principal component-based dimension reduction into the canonical correlation analysis (CCA) framework, thereby stabilizing classical test statistics in high-dimensional settings. Depending on the reduced dimension, we develop both a regularized likelihood ratio test and a regularized largest-root test to accommodate different testing scenarios. We establish the asymptotic behavior of the proposed procedures under both the null hypothesis and representative alternatives, and further develop a data-driven method for selecting the regularization parameter. Extensive simulation studies demonstrate favorable finite-sample performance across a broad range of settings.

Figures

Figures reproduced from arXiv: 2607.10500 by the authors.

Figure 7.1
Figure 7.1. Size-adjusted empirical power under the setting of [PITH_FULL_IMAGE:figures/full_fig_p022_7_1.png] view at source ↗
Figure 7.2
Figure 7.2. Same as Figure [PITH_FULL_IMAGE:figures/full_fig_p023_7_2.png] view at source ↗
Figure 7.3
Figure 7.3. Same as Figure [PITH_FULL_IMAGE:figures/full_fig_p023_7_3.png] view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

161 extracted references · 6 linked inside Pith

  1. [1]

    An Adaptable Generalization of

    Li, Haoran and Aue, Alexander and Paul, Debashis and Peng, Jie and Wang, Pei , journal=. An Adaptable Generalization of

  2. [2]

    The Annals of Probability , volume=

    No eigenvalues outside the support of the limiting spectral distribution of large-dimensional sample covariance matrices , author=. The Annals of Probability , volume=. 1998 , publisher=

    1998

  3. [3]

    The Annals of Statistics , year=

    Spectrum estimation for large dimensional covariance matrices using random matrix theory , author=. The Annals of Statistics , year=

  4. [4]

    Geometric sensitivity of random matrix results: consequences for shrinkage estimators of covariance and related statistical methods , journal =

    El Karoui, Noureddine and K. Geometric sensitivity of random matrix results: consequences for shrinkage estimators of covariance and related statistical methods , journal =

  5. [5]

    The Annals of Applied Probability , year =

    Lee, Ji Oon and Schnelli, Kevin , title =. The Annals of Applied Probability , year =

  6. [6]

    Bernoulli , year =

    Han, Xiao and Pan, Guangming and Yang, Qing , title =. Bernoulli , year =

  7. [7]

    The Annals of Statistics , volume=

    Testing in high-dimensional spiked models , author=. The Annals of Statistics , volume=

  8. [8]

    The Annals of Mathematical Statistics , year =

    James, Alan T , title =. The Annals of Mathematical Statistics , year =

Show all 161 references
  1. [9]

    2009 , author=

    Aspects of multivariate statistical theory , publisher =. 2009 , author=

    2009

  2. [10]

    1958 , publisher=

    An introduction to multivariate statistical analysis , author=. 1958 , publisher=

    1958

  3. [11]

    Bernoulli , volume=

    High-dimensional general linear hypothesis tests via non-linear spectral shrinkage , author=. Bernoulli , volume=. 2020 , publisher=

    2020

  4. [12]

    Journal of Approximation Theory , year =

    Geronimo, Jeffrey S and Hill, Theodore P , title =. Journal of Approximation Theory , year =

  5. [13]

    Computational Statistics & Data Analysis , volume=

    Practical methods for computing power in testing the multivariate general linear hypothesis , author=. Computational Statistics & Data Analysis , volume=. 1984 , publisher=

    1984

  6. [14]

    On the distribution of Roy's largest root test in

    Nadler, Boaz and Johnstone, Iain , year=. On the distribution of Roy's largest root test in

  7. [15]

    Probability Theory and Related Fields , volume=

    Approximation of Haar distributed matrices and limiting distributions of eigenvalues of Jacobi ensembles , author=. Probability Theory and Related Fields , volume=. 2009 , publisher=

    2009

  8. [16]

    Statistics and Computing , volume=

    Exact distributional computations for Roy’s statistic and the largest eigenvalue of a Wishart distribution , author=. Statistics and Computing , volume=. 2011 , publisher=

    2011

  9. [17]

    Han, Xiao and Pan, Guangming and Zhang, Bo , journal=. The. 2016 , publisher=

    2016

  10. [18]

    The annals of statistics , volume=

    Multivariate adaptive regression splines , author=. The annals of statistics , volume=. 1991 , publisher=

    1991

  11. [19]

    Journal of Multivariate Analysis , volume=

    On detection of the number of signals when the noise covariance matrix is arbitrary , author=. Journal of Multivariate Analysis , volume=. 1986 , publisher=

    1986

  12. [20]

    IEEE Journal of Selected Topics in Signal Processing , volume=

    Fundamental limit of sample generalized eigenvalue based detection of signals in noise using relatively few signal-bearing and noise-only samples , author=. IEEE Journal of Selected Topics in Signal Processing , volume=. 2010 , publisher=

    2010

  13. [21]

    Communications in Mathematical Physics , volume=

    Fredholm determinants, differential equations and matrix models , author=. Communications in Mathematical Physics , volume=. 1994 , publisher=

    1994

  14. [22]

    Level-spacing distributions and the

    Tracy, Craig A and Widom, Harold , journal=. Level-spacing distributions and the. 1994 , publisher=

    1994

  15. [23]

    Communications in Mathematical Physics , volume=

    On orthogonal and symplectic matrix ensembles , author=. Communications in Mathematical Physics , volume=. 1996 , publisher=

    1996

  16. [24]

    Johnstone and Zongming Ma and Patrick O

    Iain M. Johnstone and Zongming Ma and Patrick O. Perry and Morteza Shahram , year =

  17. [25]

    Dieng, M , journal=. A

  18. [26]

    Multivariate analysis and

    Johnstone, Iain M , journal=. Multivariate analysis and. 2008 , publisher=

    2008

  19. [27]

    Biometrika , volume=

    Power comparisons of tests of two multivariate hypotheses based on four criteria , author=. Biometrika , volume=. 1967 , publisher=

    1967

  20. [28]

    The Annals of Statistics , volume=

    Charles Stein and invariance: Beginning with the Hunt--Stein theorem , author=. The Annals of Statistics , volume=. 2021 , publisher=

    2021

  21. [29]

    Statistics , volume=

    Testing linear hypotheses in high-dimensional regressions , author=. Statistics , volume=. 2013 , publisher=

    2013

  22. [30]

    , author=

    Universality for the largest eigenvalue of sample covariance matrices with general population. , author=. The Annals of Statistics , volume=

  23. [31]

    Journal of the American Statistical Association , number=

    Testing general linear hypotheses under a high-dimensional multivariate regression model with spiked noise covariance , author=. Journal of the American Statistical Association , number=. 2023 , publisher=

    2023

  24. [32]

    El Karoui, Noureddine , journal=

  25. [33]

    Statistica Sinica , volume=

    Likelihood ratio test in multivariate linear regression: From low to high dimension , author=. Statistica Sinica , volume=. 2021 , publisher=

    2021

  26. [34]

    Rigidity of eigenvalues of generalized

    Erd. Rigidity of eigenvalues of generalized. Advances in Mathematics , volume=. 2012 , publisher=

    2012

  27. [35]

    The Annals of Applied Probability , volume=

    Central limit theorem for hotelling's T 2 statistic under large dimension , author=. The Annals of Applied Probability , volume=. 2011 , publisher=

    2011

  28. [36]

    Australian & New Zealand Journal of Statistics , volume=

    On estimation of the population spectral distribution from a high-dimensional sample covariance matrix , author=. Australian & New Zealand Journal of Statistics , volume=. 2010 , publisher=

    2010

  29. [37]

    Statistica Sinica , volume=

    Effect of high dimension: by an example of a two sample problem , author=. Statistica Sinica , volume=. 1996 , publisher=

    1996

  30. [38]

    2004 , publisher=

    Bai, Zhidong and Silverstein, Jack W , journal=. 2004 , publisher=

    2004

  31. [39]

    1991 , publisher=

    Probability metrics and the stability of stochastic models , author=. 1991 , publisher=

    1991

  32. [40]

    Journal of Multivariate Analysis , volume=

    A nonparametric two-sample test applicable to high dimensional data , author=. Journal of Multivariate Analysis , volume=. 2014 , publisher=

    2014

  33. [41]

    Journal of the Royal Statistical Society: Series B (Statistical Methodology) , volume=

    Two-sample test of high dimensional means under dependence , author=. Journal of the Royal Statistical Society: Series B (Statistical Methodology) , volume=. 2014 , publisher=

    2014

  34. [42]

    High-dimensional sparse

    Cai, T Tony and Xia, Yin , journal=. High-dimensional sparse. 2014 , publisher=

    2014

  35. [43]

    Biometrics , volume=

    Simulation-based hypothesis testing of high dimensional means under covariance heterogeneity , author=. Biometrics , volume=. 2017 , publisher=

    2017

  36. [44]

    The Annals of Statistics , volume=

    A two-sample test for high-dimensional data with applications to gene-set testing , author=. The Annals of Statistics , volume=. 2010 , publisher=

    2010

  37. [45]

    A regularized

    Chen, Lin S and Paul, Debashis and Prentice, Ross L and Wang, Pei , journal=. A regularized. 2011 , publisher=

    2011

  38. [46]

    arXiv preprint arXiv:1410.2848 , year=

    Two-sample tests for high dimensional means with thresholding and data transformation , author=. arXiv preprint arXiv:1410.2848 , year=

    arXiv

  39. [47]

    2000 , publisher=

    Probability and Measure Theory , author=. 2000 , publisher=

    2000

  40. [48]

    The Annals of Mathematical Statistics , pages=

    A high dimensional two sample significance test , author=. The Annals of Mathematical Statistics , pages=. 1958 , publisher=

    1958

  41. [49]

    Biometrics , volume=

    A significance test for the separation of two highly multivariate small samples , author=. Biometrics , volume=. 1960 , publisher=

    1960

  42. [50]

    Journal of Multivariate Analysis , volume=

    Shrinkage-based diagonal Hotelling's tests for high-dimensional small sample size data , author=. Journal of Multivariate Analysis , volume=. 2016 , publisher=

    2016

  43. [51]

    Journal of the American Statistical Association , volume=

    Multivariate-sign-based high-dimensional tests for the two-sample location problem , author=. Journal of the American Statistical Association , volume=. 2016 , publisher=

    2016

  44. [52]

    Journal of the Korean Statistical Society , volume=

    Multivariate analysis for the case when the dimension is large compared to the sample size , author=. Journal of the Korean Statistical Society , volume=

  45. [53]

    Asymptotic results of a high dimensional

    Fujikoshi, Yasunori and Himeno, Tetsuto and Wakaki, Hirofumi , journal=. Asymptotic results of a high dimensional. 2004 , publisher=

    2004

  46. [54]

    Journal of the Royal Statistical Society: Series B (Statistical Methodology) , volume=

    Tests for high dimensional generalized linear models , author=. Journal of the Royal Statistical Society: Series B (Statistical Methodology) , volume=. 2016 , publisher=

    2016

  47. [55]

    Annals of the Institute of Statistical Mathematics , volume=

    On testing the equality of high dimensional mean vectors with unequal covariance matrices , author=. Annals of the Institute of Statistical Mathematics , volume=. 2017 , publisher=

    2017

  48. [56]

    Journal of Multivariate Analysis , volume=

    A well-conditioned estimator for large-dimensional covariance matrices , author=. Journal of Multivariate Analysis , volume=. 2004 , publisher=

    2004

  49. [57]

    The Annals of Statistics , volume=

    Nonlinear shrinkage estimation of large-dimensional covariance matrices , author=. The Annals of Statistics , volume=. 2012 , publisher=

    2012

  50. [58]

    Liu, Haoyang and Aue, Alexander and Paul, Debashis , journal=. On the. 2015 , publisher=

    2015

  51. [59]

    Advances in Neural Information Processing Systems , pages=

    A more powerful two-sample test in high dimensions using random projection , author=. Advances in Neural Information Processing Systems , pages=

  52. [60]

    2013 , publisher=

    Concentration inequalities: A nonasymptotic theory of independence , author=. 2013 , publisher=

    2013

  53. [61]

    Central limit theorem for

    Pan, Guangming and Zhou, Wang , journal=. Central limit theorem for. 2011 , publisher=

    2011

  54. [62]

    Journal of Statistical Planning and Inference , volume=

    A test for the mean vector in large dimension and small samples , author=. Journal of Statistical Planning and Inference , volume=. 2013 , publisher=

    2013

  55. [63]

    Journal of Statistical Planning and Inference , volume=

    Random matrix theory in statistics: A review , author=. Journal of Statistical Planning and Inference , volume=. 2014 , publisher=

    2014

  56. [64]

    Journal of Multivariate Analysis , volume=

    On the empirical distribution of eigenvalues of a class of large dimensional random matrices , author=. Journal of Multivariate Analysis , volume=. 1995 , publisher=

    1995

  57. [65]

    Journal of Multivariate Analysis , volume=

    A test for the mean vector with fewer observations than the dimension , author=. Journal of Multivariate Analysis , volume=. 2008 , publisher=

    2008

  58. [66]

    Journal of Multivariate Analysis , volume=

    A test for the mean vector with fewer observations than the dimension under non-normality , author=. Journal of Multivariate Analysis , volume=. 2009 , publisher=

    2009

  59. [67]

    Journal of Multivariate Analysis , volume=

    A two sample test in high dimensional data , author=. Journal of Multivariate Analysis , volume=. 2013 , publisher=

    2013

  60. [68]

    2016 , publisher=

    Srivastava, Radhendushka and Li, Ping and Ruppert, David , journal=. 2016 , publisher=

    2016

  61. [69]

    arXiv preprint arXiv:1011.3027 , year=

    Introduction to the non-asymptotic analysis of random matrices , author=. arXiv preprint arXiv:1011.3027 , year=

    arXiv

  62. [70]

    An adaptable generalization of

    Li, Haoran and Aue, Alexander and Paul, Debashis and Peng, Jie and Wang, Pei , journal=. An adaptable generalization of

  63. [71]

    Statistica Sinica , pages=

    Jackknife empirical likelihood test for equality of two high dimensional means , author=. Statistica Sinica , pages=. 2013 , publisher=

    2013

  64. [72]

    Biometrika , volume=

    An adaptive two-sample test for high-dimensional means , author=. Biometrika , volume=. 2016 , publisher=

    2016

  65. [73]

    Probability Theory and Related Fields , volume=

    On the limit of the largest eigenvalue of the large dimensional sample covariance matrix , author=. Probability Theory and Related Fields , volume=. 1988 , publisher=

    1988

  66. [74]

    Journal of Multivariate Analysis , volume=

    Testing homogeneity of mean vectors under heteroscedasticity in high-dimension , author=. Journal of Multivariate Analysis , volume=. 2015 , publisher=

    2015

  67. [75]

    Journal of Multivariate Analysis , volume=

    Multivariate analysis of variance with fewer observations than the dimension , author=. Journal of Multivariate Analysis , volume=. 2006 , publisher=

    2006

  68. [76]

    Journal of the American Statistical Association , volume=

    A high-dimensional nonparametric multivariate test for mean vector , author=. Journal of the American Statistical Association , volume=. 2015 , publisher=

    2015

  69. [77]

    Probability Theory and Related Fields , volume=

    Eigenvectors of some large sample covariance matrix ensembles , author=. Probability Theory and Related Fields , volume=. 2011 , publisher=

    2011

  70. [78]

    Journal of Multivariate Analysis , volume=

    Strong convergence of the empirical distribution of eigenvalues of large dimensional random matrices , author=. Journal of Multivariate Analysis , volume=. 1995 , publisher=

    1995

  71. [79]

    Journal of Multivariate Analysis , volume=

    Analysis of the limiting spectral distribution of large dimensional random matrices , author=. Journal of Multivariate Analysis , volume=. 1995 , publisher=

    1995

  72. [80]

    The Annals of Probability , pages=

    Exact separation of eigenvalues of large dimensional sample covariance matrices , author=. The Annals of Probability , pages=. 1999 , publisher=

    1999

  73. [81]

    Probability and Measure , author=

  74. [82]

    Convergence of Probability Measures , author=

  75. [83]

    Proceedings of the National Academy of Sciences , volume=

    Maximum properties and inequalities for the eigenvalues of completely continuous operators , author=. Proceedings of the National Academy of Sciences , volume=. 1951 , publisher=

    1951

  76. [84]

    Journal of Statistical Planning and Inference , volume=

    High-dimensional general linear hypothesis testing under heteroscedasticity , author=. Journal of Statistical Planning and Inference , volume=. 2017 , publisher=

    2017

  77. [85]

    Statistica Sinica , pages=

    Asymptotics of sample eigenstructure for a large dimensional spiked covariance model , author=. Statistica Sinica , pages=. 2007 , publisher=

    2007

  78. [86]

    Limiting behavior of eigenvalues in high-dimensional

    Bai, Zhidong and Choi, Kwok Pui and Fujikoshi, Yasunori , year=. Limiting behavior of eigenvalues in high-dimensional

  79. [87]

    Journal of Multivariate Analysis , volume=

    No eigenvalues outside the support of the limiting empirical spectral distribution of a separable covariance matrix , author=. Journal of Multivariate Analysis , volume=. 2009 , publisher=

    2009

  80. [88]

    The Annals of Statistics , volume=

    Universality for the largest eigenvalue of sample covariance matrices with general population , author=. The Annals of Statistics , volume=. 2015 , publisher=

    2015

  81. [89]

    Central limit theorems for linear spectral statistics of large dimensional

    Zheng, Shurong , booktitle=. Central limit theorems for linear spectral statistics of large dimensional

  82. [90]

    2017 , publisher=

    Zheng, Shurong and Bai, Zhidong and Yao, Jianfeng , journal=. 2017 , publisher=

    2017

  83. [91]

    The Annals of Statistics , volume=

    Testing for independence of large dimensional vectors , author=. The Annals of Statistics , volume=. 2019 , publisher=

    2019

  84. [92]

    Journal of the American Statistical Association , volume=

    Linear hypothesis testing in dense high-dimensional linear models , author=. Journal of the American Statistical Association , volume=. 2018 , publisher=

    2018

  85. [93]

    IEEE Transactions on Signal Processing , volume=

    Signal detection via spectral theory of large dimensional random matrices , author=. IEEE Transactions on Signal Processing , volume=. 1992 , publisher=

    1992

  86. [94]

    2010 , publisher=

    Spectral analysis of large dimensional random matrices , author=. 2010 , publisher=

    2010

  87. [95]

    Matematicheskii Sbornik , volume=

    Distribution of eigenvalues for some sets of random matrices , author=. Matematicheskii Sbornik , volume=

  88. [96]

    Random Matrices: Theory and Applications , volume =

    Li, Haoran , title =. Random Matrices: Theory and Applications , volume =

  89. [97]

    Probability Theory and Related Fields , volume=

    Anisotropic local laws for random matrices , author=. Probability Theory and Related Fields , volume=. 2017 , publisher=

    2017

  90. [98]

    Probability Theory and Related Fields , volume =

    Bao, Zhigang and Pan, Guangming and Zhou, Wang , title =. Probability Theory and Related Fields , volume =. 2013 , doi =

    2013

  91. [99]

    Isotropic local laws for sample covariance and generalized

    Bloemendal, Alex and Erdős, László and Knowles, Antti and Yau, Horng‑Tzer and Yin, Jun , journal =. Isotropic local laws for sample covariance and generalized. 2014 , doi =

    2014

  92. [100]

    Electronic Journal of Probability , volume=

    The local semicircle law for a general class of random matrices , author=. Electronic Journal of Probability , volume=. 2013 , publisher=

    2013

  93. [101]

    Neuroimage , volume=

    The WU-Minn human connectome project: an overview , author=. Neuroimage , volume=. 2013 , publisher=

    2013

  94. [102]

    Journal of the American Statistical Association , volume=

    Testing General Linear Hypotheses Under a High-Dimensional Multivariate Regression Model with Spiked Noise Covariance , author=. Journal of the American Statistical Association , volume=. 2024 , publisher=

    2024

  95. [103]

    Proximity of probability distributions in terms of

    Bobkov, Sergei Germanovich , journal=. Proximity of probability distributions in terms of. 2016 , publisher=

    2016

  96. [104]

    Journal of Multivariate Analysis , volume=

    An overview of tests on high-dimensional means , author=. Journal of Multivariate Analysis , volume=. 2022 , publisher=

    2022

  97. [105]

    Journal of the American Statistical Association , volume=

    Linear hypothesis testing in linear models with high-dimensional responses , author=. Journal of the American Statistical Association , volume=. 2022 , publisher=

    2022

  98. [106]

    Journal of the American Statistical Association , volume=

    Projection test for mean vector in high dimensions , author=. Journal of the American Statistical Association , volume=. 2024 , publisher=

    2024

  99. [107]

    , title =

    Johnstone, Iain M. , title =. Annals of Statistics , volume =

  100. [108]

    Annals of Applied Probability , volume =

    Onatski, Alexei , title =. Annals of Applied Probability , volume =

  101. [109]

    Acta Mathematica , volume =

    Tao, Terence and Vu, Van , title =. Acta Mathematica , volume =

  102. [110]

    Bulk universality for generalized

    Erd. Bulk universality for generalized. Probability Theory and Related Fields , volume =

  103. [111]

    Duke Mathematical Journal , volume =

    Lee, Ji Oon and Yin, Jun , title =. Duke Mathematical Journal , volume =

  104. [112]

    and Yin, Jun , title =

    Pillai, Natesh S. and Yin, Jun , title =. Annals of Statistics , volume =

  105. [113]

    and Yin, Jun , title =

    Pillai, Natesh S. and Yin, Jun , title =. Annals of Applied Probability , volume =

  106. [114]

    Annals of Statistics , volume =

    Bao, Zhigang and Pan, Guangming and Zhou, Wang , title =. Annals of Statistics , volume =

  107. [115]

    Reviews in Mathematical Physics , volume =

    Lee, Ji Oon and Schnelli, Konstanze , title =. Reviews in Mathematical Physics , volume =

  108. [116]

    The Annals of Applied Probability , volume =

    Ding, Xiucai and Yang, Fan , title =. The Annals of Applied Probability , volume =. 2018 , doi =

    2018

  109. [117]

    Electronic Journal of Probability , volume =

    Yang, Fan , title =. Electronic Journal of Probability , volume =. 2019 , doi =

    2019

  110. [118]

    IEEE Transactions on Information Theory , volume =

    Ding, Xiucai and Yang, Fan , title =. IEEE Transactions on Information Theory , volume =. 2022 , doi =

    2022

  111. [119]

    arXiv preprint arXiv:2008.04166 , volume=

    Tracy-Widom distribution for the edge eigenvalues of Gram type random matrices , author=. arXiv preprint arXiv:2008.04166 , volume=

    2008 arXiv

  112. [120]

    Information and Inference: A Journal of the IMA , volume =

    Ding, Xiucai and Xie, Jiahui , title =. Information and Inference: A Journal of the IMA , volume =. 2025 , doi =

    2025

  113. [121]

    arXiv preprint arXiv:2404.14751 , year=

    Eigenvector distributions and optimal shrinkage estimators for large covariance and precision matrices , author=. arXiv preprint arXiv:2404.14751 , year=

    arXiv

  114. [122]

    Electronic Journal of Probability , volume =

    Yang, Fan , title =. Electronic Journal of Probability , volume =. 2019 , pages =

    2019

  115. [123]

    The Annals of Applied Probability , volume =

    Hwang, Jong Yun and Lee, Ji Oon and Schnelli, Konstanze , title =. The Annals of Applied Probability , volume =. 2022 , doi =

    2022

  116. [124]

    arXiv preprint arXiv:2303.03532 , year=

    Extreme eigenvalues of sample covariance matrices under generalized elliptical models with applications , author=. arXiv preprint arXiv:2303.03532 , year=

    arXiv

  117. [125]

    Random Matrices: Theory and Applications , volume=

    Spiked sample covariance matrices with possibly multiple bulk components , author=. Random Matrices: Theory and Applications , volume=. 2021 , publisher=

    2021

  118. [126]

    The Annals of Statistics , volume=

    SPIKED SEPARABLE COVARIANCE MATRICES AND PRINCIPAL COMPONENTS , author=. The Annals of Statistics , volume=. 2021 , publisher=

    2021

  119. [127]

    Information and Inference: A Journal of the IMA , volume=

    Tracy--Widom distribution for the edge eigenvalues of elliptical model , author=. Information and Inference: A Journal of the IMA , volume=. 2025 , publisher=

    2025

  120. [128]

    Phase Transition of the Largest Eigenvalue for Nonnull Complex Sample Covariance Matrices , journal =

    Baik, Jinho and Ben Arous, G. Phase Transition of the Largest Eigenvalue for Nonnull Complex Sample Covariance Matrices , journal =. 2005 , doi =

    2005

  121. [129]

    , title =

    Baik, Jinho and Silverstein, Jack W. , title =. Journal of Multivariate Analysis , volume =. 2006 , doi =

    2006

  122. [130]

    Journal of Multivariate Analysis , volume=

    Edge statistics of large dimensional deformed rectangular matrices , author=. Journal of Multivariate Analysis , volume=. 2022 , publisher=

    2022

  123. [131]

    arXiv preprint arXiv:2403.15792 , year=

    Reviving pseudo-inverses: Asymptotic properties of large dimensional Moore-Penrose and Ridge-type inverses with applications , author=. arXiv preprint arXiv:2403.15792 , year=

  124. [132]

    Bernoulli , volume =

    Zhixiang Zhang and Yiming Liu and Guangming Pan , title =. Bernoulli , volume =. 2024 , doi =

    2024

  125. [133]

    Journal of Multivariate Analysis , volume=

    On the empirical distribution of eigenvalues of large dimensional information-plus-noise-type matrices , author=. Journal of Multivariate Analysis , volume=. 2007 , publisher=

    2007

  126. [134]

    Journal of Multivariate Analysis , volume=

    Analysis of the limiting spectral distribution of large dimensional information-plus-noise type matrices , author=. Journal of Multivariate Analysis , volume=. 2007 , publisher=

    2007

  127. [135]

    Science China Mathematics , volume=

    Limiting spectral distribution of high-dimensional noncentral fisher matrices and its analysis , author=. Science China Mathematics , volume=. 2023 , publisher=

    2023

  128. [136]

    Computational Statistics & Data Analysis , volume=

    Numerical implementation of the QuEST function , author=. Computational Statistics & Data Analysis , volume=. 2017 , publisher=

    2017

  129. [137]

    Spectrum estimation: A unified framework for covariance matrix estimation and

    Ledoit, Olivier and Wolf, Michael , journal=. Spectrum estimation: A unified framework for covariance matrix estimation and. 2015 , publisher=

    2015

  130. [138]

    Journal of Statistical Planning and Inference , volume=

    Estimation of the population spectral distribution from a large dimensional sample covariance matrix , author=. Journal of Statistical Planning and Inference , volume=. 2013 , publisher=

    2013

  131. [139]

    Journal of the American Statistical Association , volume=

    Block-diagonal covariance selection for high-dimensional Gaussian graphical models , author=. Journal of the American Statistical Association , volume=. 2018 , publisher=

    2018

  132. [140]

    Journal of Multivariate Analysis , volume=

    A nonparametric test for block-diagonal covariance structure in high dimension and small samples , author=. Journal of Multivariate Analysis , volume=. 2019 , publisher=

    2019

  133. [141]

    Annals of Statistics , volume=

    Independence test for high dimensional data based on regularized canonical correlation coefficients , author=. Annals of Statistics , volume=. 2015 , publisher=

    2015

  134. [142]

    British Journal of Statistical Psychology , volume=

    The relations of the newer multivariate statistical methods to factor analysis , author=. British Journal of Statistical Psychology , volume=. 1957 , publisher=

    1957

  135. [143]

    1957 , publisher=

    A course in multivariate analysis , author=. 1957 , publisher=

    1957

  136. [144]

    arXiv preprint arXiv:2504.15510 , year=

    Ridge-Regularized Largest Root Test For High-Dimensional General Linear Hypotheses , author=. arXiv preprint arXiv:2504.15510 , year=

    arXiv

  137. [145]

    The Annals of Statistics , volume=

    Canonical correlation coefficients of high-dimensional Gaussian vectors: Finite rank case , author=. The Annals of Statistics , volume=. 2019 , publisher=

    2019

  138. [146]

    The Annals of Statistics , volume=

    CENTRAL LIMIT THEOREMS FOR CLASSICAL LIKELIHOOD RATIO TESTS FOR HIGH-DIMENSIONAL NORMAL DISTRIBUTIONS , author=. The Annals of Statistics , volume=

  139. [147]

    Electronic Journal of Statistics , volume=

    Test of independence for high-dimensional random vectors based on freeness in block correlation matrices , author=. Electronic Journal of Statistics , volume=

  140. [148]

    Statistica Neerlandica , volume=

    Testing block-diagonal covariance structure for high-dimensional data , author=. Statistica Neerlandica , volume=. 2015 , publisher=

    2015

  141. [149]

    Science China Mathematics , volume=

    Testing the independence of sets of large-dimensional variables , author=. Science China Mathematics , volume=. 2013 , publisher=

    2013

  142. [150]

    Journal of Multivariate Analysis , volume=

    Testing block-diagonal covariance structure for high-dimensional data under non-normality , author=. Journal of Multivariate Analysis , volume=. 2017 , publisher=

    2017

  143. [151]

    The Annals of Applied Statistics , volume=

    Approximate null distribution of the largest root in multivariate analysis , author=. The Annals of Applied Statistics , volume=

  144. [152]

    Methodology and Applications of Statistics: A Volume in Honor of CR Rao on the Occasion of his 100th Birthday , pages=

    Limiting Canonical Distribution of Two Large-Dimensional Random Vectors , author=. Methodology and Applications of Statistics: A Volume in Honor of CR Rao on the Occasion of his 100th Birthday , pages=. 2022 , publisher=

    2022

  145. [153]

    The Annals of Statistics , volume=

    MEASURING AND TESTING DEPENDENCE BY CORRELATION OF DISTANCES , author=. The Annals of Statistics , volume=

  146. [154]

    The Annals of Statistics , volume=

    Hypothesis testing on linear structures of high dimensional covariance matrix , author=. The Annals of Statistics , volume=

  147. [155]

    The Journal of Machine Learning Research , volume=

    Consistent nonparametric tests of independence , author=. The Journal of Machine Learning Research , volume=. 2010 , publisher=

    2010

  148. [156]

    Journal of Machine Learning Research , volume=

    Kernel Methods for Measuring Independence , author=. Journal of Machine Learning Research , volume=

  149. [157]

    transformation , volume=

    High-Dimensional Independence Testing via Maximum and Average Distance Correlations , author=. transformation , volume=

  150. [158]

    Scandinavian Journal of Statistics , year=

    Testing independence between high-dimensional random vectors using rank-based max-sum tests , author=. Scandinavian Journal of Statistics , year=

  151. [159]

    Communications in Statistics-Theory and Methods , volume=

    Testing for independence of sets of high-dimensional normal vectors using random projection approach , author=. Communications in Statistics-Theory and Methods , volume=. 2025 , publisher=

    2025

  152. [160]

    TEST , volume=

    Block-diagonal test for high-dimensional covariance matrices , author=. TEST , volume=. 2023 , publisher=

    2023

  153. [161]

    Journal of Multivariate Analysis , volume=

    Multivariate tests of independence and their application in correlation analysis between financial markets , author=. Journal of Multivariate Analysis , volume=. 2020 , publisher=

    2020

Pith tools

Reviewed July 14, 2026 · model on record in the stance chip above.