Hypothesis Testing for High-Dimensional Matrix-Valued Data

Danning Li; Lingzhou Xue; Runze Li; Shijie Cui

REVIEW 4 major objections 4 minor 1 cited by

Hypothesis Testing for High-Dimensional Matrix-Valued Data

T0 review · 4 major / 4 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read The paper proposes a U-statistic difference test for the rank of a high-dimensional matrix mean, proves asymptotic normality of its oracle and plug-in versions, and demonstrates the test on surveillance-video object detection.

desk verdict A sensible rank test for high-dimensional matrix means that is undermined by an over-broad oracle theorem, but the underlying idea is worth a serious referee. read the letter →

arxiv 2412.07987 v1 pith:ESDTVD4N submitted 2024-12-10 stat.ME math.STstat.MLstat.TH

Shijie Cui , Danning Li , Runze Li , Lingzhou Xue This is my paper · ORCID

classification stat.MEmath.STstat.MLstat.TH MSC 62H1562F0562G2062H25

keywords matrix-valueddatarankhypothesistestinghigh-dimensionalstatisticsU-statisticssparsesingularvaluedecompositionminimumdiscrepancytestsurveillancevideoanalysis

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

This paper tries to establish a valid test for whether the mean matrix of high-dimensional matrix-valued data has rank at most $K$, in the regime where the matrix dimensions $q$ and $p$ grow with or exceed the sample size $n$. The classical minimum discrepancy test works only when $\max(q,p)=o(n^{1/4})$ and loses size control in high dimensions, so the authors build a different statistic: compare an estimate of the total squared signal in the mean matrix with an estimate of the sum of the $K$ largest squared singular values. Under the null these two quantities agree in expectation, and the paper proves that their difference, after scaling by an estimate of the noise covariance trace, is asymptotically standard normal, with known local power. The result matters because rank is a structural property of images and other array data, and this is the first test designed for that question when dimensions are large. The paper also supplies the plug-in machinery, a sparse SVD estimator for singular vectors and a ratio-consistent estimator of $\mathrm{tr}(\Sigma^2)$, and validates the method on surveillance videos.

What carries the argument

The load-bearing object is the difference of two U-statistics: $U_n = \mathrm{tr}\bigl(\tfrac{1}{n(n-1)}\sum_{i\ne j}X_iX_j^\top\bigr)$ estimates the total squared Frobenius norm of the mean, and $T_k = \tfrac{1}{n(n-1)}\sum_{i\ne j}\mu_k^\top X_i V_0V_0^\top X_j^\top \mu_k$ estimates the $k$-th squared singular value using the oracle singular vectors. The plug-in statistic $\hat G_n = (U_n - \sum_{k=1}^K \hat T_k)/\sqrt{2\hat S^2_{n1}/n^2}$ is built from a sparse SVD estimator that solves the penalized Frobenius problem (3.7) with a group concave penalty, and from a ratio-consistent estimator $\hat S^2_{n1}$ of $\mathrm{tr}(\Sigma^2)$. The theoretical work is to show that the sparse SVD error is small enough (A10) that the plug-in statistic inherits the oracle normal limit, while the minimum discrepancy statistic fails when dimensions grow.

What would settle it

Take $q=p=K$, $\Pi_0=0$, and $\Sigma=I_{qp}$. Then with any full-rank oracle pair $U_0,V_0$ the statistic $V_n = U_n - \sum_{k=1}^K T_k$ is identically zero for every sample, while the claimed scaling $2\mathrm{tr}(\Sigma^2)/n^2 = 2p^2/n^2$ is positive, so the central limit theorem cannot hold. A concrete check is to simulate $V_n$ in this setting, estimate its variance, and watch it vanish as $K$ approaches $\min(q,p)$.

Watch

Extended reading notes

Core claim

The central claim is that the oracle statistic $V_n = U_n - \sum_{k=1}^K T_k$, where $U_n$ estimates $\mathrm{tr}(\Pi_0\Pi_0^\top)$ and $T_k$ estimates the $k$-th squared singular value $\sigma_k^2$, is asymptotically normal with variance $2\mathrm{tr}(\Sigma^2)/n^2$ under $H_0: \mathrm{rank}(\Pi_0)\le K$, and has nontrivial power when $n\sum_{i>K}\sigma_i^2/\sqrt{2\mathrm{tr}(\Sigma^2)}\to\delta$. The sample version $\hat G_n$ replaces unknown singular vectors by a sparse SVD estimator and replaces $\mathrm{tr}(\Sigma^2)$ by a ratio-consistent estimator; under assumptions A1, A4, A9 and A10 the sample version has the same null limit and the same local power. Thus the paper asserts that rank of a high-dimensional matrix mean is testable with standard normal p-values, without estimating the inverse covariance matrix.

Load-bearing premise

The paper assumes that subtracting the estimated top-$K$ squared singular-value contributions does not change the variability of the test statistic, so that the normal approximation with variance $2\mathrm{tr}(\Sigma^2)/n^2$ is valid; this assumption is stated nowhere, and it fails when $K$ is close to the matrix dimensions.

Editorial extensions

If this is right

A practitioner can test $H_0: \mathrm{rank}(\Pi_0)\le K$ with standard normal critical values even when $q$ and $p$ are much larger than $n$, a regime where the minimum discrepancy statistic's $\chi^2$ approximation breaks down.
Sequential testing over $K=0,1,\ldots$ yields an estimate of the true rank, and the paper shows this estimate tracks the presence of pedestrians and cars in two surveillance-video datasets.
Under the sparsity assumptions, the sparse SVD estimator recovers the row and column support of the singular vectors at rate $(\sqrt{k}+\sqrt{l})/\sqrt{n}$, so the plug-in test is valid whenever $n(k+l)=o(\mathrm{tr}(\Sigma^2))$.
Setting $K=0$ reduces the new statistic to the L2-type high-dimensional mean test from which the U-statistic construction is drawn, making the matrix rank test a generalization of an existing vector test.
The local power formula shows the test can detect alternatives with signal $\delta = n\sum_{i>K}\sigma_i^2/\sqrt{2\mathrm{tr}(\Sigma^2)}$, and that power drops predictably as the noise covariance trace grows.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

As an extension the paper does not state, the normal calibration requires $K$ to be small relative to $\min(q,p)$; at $K=\min(q,p)$ with a full-rank projection, $V_n$ can be identically zero, so a corrected theorem would either restrict $K$ or compute the variance of the subtracted part explicitly.
The tuning-parameter selection by sample splitting is a practical bottleneck; the theory assumes a well-behaved penalty, so a fully data-driven choice (for example an analog of BIC under non-identity noise) would be the natural next development.
The same difference-of-sums logic could be carried to tensor-valued data by replacing squared singular values with Tucker-core norms, although the paper mentions tensors only as motivation.
The video experiments suggest estimated rank could serve as a continuous event signal for counting or sizing objects; the paper itself reports only that rank rises when objects enter and falls when they leave.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

The paper takes a real problem — rank testing for the mean of matrix-valued data when dimensions grow with n — and proposes a natural solution: a U-statistic designed to separate the top-K singular value contribution from the rest, plugged with a sparse SVD estimator. The motivation is solid, and the paper correctly shows that the classic Cragg–Donald minimum discrepancy test breaks down as dimensions increase. The oracle statistic is a straightforward but sensible adaptation of Chen et al. (2010), and the sparse SVD theory is a credible extension of Lee et al. (2010). The simulations and the two video case studies are also a plus, demonstrating that the method works in practice for K=1 and that the minimum discrepancy test does not control size.

The soft spot is load-bearing. Theorem 3, as stated, claims the oracle statistic is asymptotically N(0,1) under H0 for all K, with variance 2 tr(Sigma^2)/n^2. That is false. The variance calculation ignores the projection term Q = I - P_V ⊗ P_U. The stress-test example is correct: with q=p=n, Sigma=I, and K around n/2, the asymptotic variance is 3/4 of the claimed denominator, so the calibration fails under the paper's own assumptions. The theorem needs an explicit condition on K — something like K^2 = o(tr(Sigma^2)) or K fixed — and a rigorous variance computation. This is not a minor typo; it propagates to Theorem 5, which inherits the same gap. The simulations only exercise K=1, where the projection term is negligible, so they never expose the flaw.

That said, the paper is not sloppy in its overall approach. The sparse SVD theory, the plug-in analysis, and the empirical work are all coherent. The main missing piece is a correct statement and proof of the oracle theorem under appropriate restrictions on K. The conclusion overstates novelty slightly — claiming to be the first work on testing the structure of matrix-valued data ignores related rank tests in econometrics — but the high-dimensional matrix-mean setting is genuinely new.

My recommendation: this paper deserves a serious referee, not a desk reject. The idea is worth developing, and the flaw appears fixable. But in its current form, I would not accept it; the central asymptotic result is incorrect as stated. If the authors can supply a corrected theorem with a proper condition on K and update the plug-in theory accordingly, the paper could make a useful contribution.

Referee Report

4 major / 4 minor

Summary. The paper proposes a test for H0: rank(Π0) ≤ K for the mean of high-dimensional matrix-valued data with separable covariance. The authors first argue that the classical Cragg-Donald minimum-discrepancy test fails when matrix dimensions grow with the sample size, then introduce an oracle U-statistic that subtracts the first K estimated singular-value contributions from the total squared Frobenius norm, claim asymptotic normality under the null and local alternatives, and provide a sparse SVD estimator to make the statistic plug-in. Simulations and two surveillance-video case studies are used to support the method.

Significance. The testing problem is timely, and the oracle statistic is a natural U-statistic extension of high-dimensional mean tests to matrix rank testing. The sparse SVD estimator with concave penalties is a useful contribution, and the video surveillance applications are interesting. If the central distributional claims were valid, the paper would be a meaningful step for matrix-valued inference. However, the main theorem is not valid in the stated generality, and the plug-in theorem inherits the same gap; the paper's own limitations and the missing proofs must be addressed before the claims can be accepted.

major comments (4)

[3.2, Theorem 3] The variance of the oracle statistic is miscalibrated when K is not negligible. Writing Q = I - V0V0ᵀ ⊗ U0U0ᵀ and x_i = vec(X_i), the statistic V_n equals (1/(n(n-1))) Σ_{i≠j} x_iᵀ Q x_j. Under H0, Q vec(Π0) = 0, so this is a degenerate U-statistic with variance 2 tr(QΣQΣ)/n² + o(n⁻²), not 2 tr(Σ²)/n² as used in the denominator of (3.5). The difference tr(Σ²) − tr(QΣQΣ) = 2 tr((V0V0ᵀ⊗U0U0ᵀ)Σ²) − tr((V0V0ᵀ⊗U0U0ᵀ)Σ(V0V0ᵀ⊗U0U0ᵀ)Σ) can be of the same order as tr(Σ²) when K grows. For example, take q = p = n, Σ = I_{n²}, and Π0 of rank K = floor(n/2) with U0 = V0 equal to the first K standard basis vectors. Assumptions A1 and A4 hold, but tr(QΣQΣ) = (3/4) tr(Σ²) + o(tr(Σ²)), so G_n converges to N(0, 3/4), not N(0, 1). No condition on K appears in Theorem 3 or in A4; a condition such as K² ||Σ||₂² = o(tr(Σ²)), or an explicit variance calculation, is required.
[3.4, Theorem 5] The plug-in statistic inherits the missing variance correction from Theorem 3. Assumptions A9 and A10 impose no restriction linking K to tr(Σ²); A10 only requires the projection error to be o_P(√tr(Σ²)/n), which does not make the Q-factor in the variance vanish. Consequently the local-power statement (3.9) is also unproved in the stated generality. The simulation section (Section 4, Tables 1–3) only exercises K = 1: the null models in Model (a) with c = 0 and Model (b) with c = 0 have rank-1 mean matrices, so the subtracted projection has rank K² = 1 and the variance correction is negligible. The simulations therefore do not test the regime in which Theorem 3 fails.
[3.2 and 3.4, null with R < K] The null hypothesis H0: rank(Π0) ≤ K includes cases where the true rank R is strictly less than K. In those cases the singular vectors μ_k, ν_k for k > R are not identified, so the oracle statistic (3.4) is not a well-defined functional of the data-generating process; for general Σ, its null distribution can depend on the arbitrary completion of U0 and V0 through terms such as tr((V0V0ᵀ⊗U0U0ᵀ)Σ²). Theorem 4 explicitly assumes 0 < K ≤ R, so the estimation theory does not cover the R < K part of the null. The paper should either restrict the null to R = K or prove that the statistical behavior is invariant to the completion and that A10 is attainable in that case.
[Appendix / proofs] The manuscript states in Section 1 and elsewhere that proofs are presented in the appendix, but no appendix is included in the posted text. Since the core results are asymptotic theorems for degenerate U-statistics and a new sparse SVD estimator, the absence of the proofs makes the central claims unverifiable as submitted. This is a load-bearing omission for a theory-focused paper.

minor comments (4)

[3.1, Theorem 2] The display for Theorem 2 appears to be missing a square root in the denominator; it should be (T_R − (q−R)(p−R)) / √(2(q−R)(p−R)) → N(0, 1), not the expression with 2(q−R)(p−R) in the denominator as printed.
[3.3, Assumption A8] The condition for λ_v uses max_{v∈Nu0}, which should be max_{v∈Nv0}; the two displayed conditions for λ_u and λ_v otherwise appear identical, suggesting a copy-paste error.
[2.2 and 4] There are typos including 'subgaussion' for 'subgaussian', 'Euclidan' for 'Euclidean', and 'can ’t' in Remark 1; the label 'REMARK 1' in Table 3 should be replaced by a proper model description.
[3.4, estimator of tr(Σ²)] The displayed estimator \S\S^2_{n1} is garbled: the first term as printed, tr(2XiX T j )2, should presumably be tr(X_i X_jᵀ)² with the correct combinatorial factors; please check the formula against Li and Chen (2012).

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the statistic and null calibration are derived from U-statistic theory, with self-citations only in background; the known variance-normalization gap is a correctness issue, not a circular reduction.

full rationale

I walked the claimed derivation chain. The oracle statistic V_n = U_n - sum_{k=1}^K T_k is a U-statistic whose null mean is zero by the rank-K structure of Pi_0, and the claimed N(0,1) calibration is obtained from standard degenerate U-statistic CLT reasoning under A1 and A4; no parameter is fitted to the null. The plug-in Theorem 5 is conditional on explicit convergence-rate assumptions A9-A10 on the sparse SVD estimators, and Theorem 4 independently establishes such rates under A5-A8. The variance estimator ^S^2_{n1} follows Li & Chen (2012), an external source. Self-citations (Yu et al. 2023, Li et al. 2024, Huang et al. 2022) appear in the literature review and are not load-bearing for Theorems 3 or 5. The main caveat is a mathematical gap rather than circularity: Theorem 3 states the variance as 2 tr(Sigma_2)/n^2 without restricting K or accounting for the projection I - V0V0^T tensor U0U0^T, so the normalization can fail for large K; this is an omitted-condition error, not an equivalence between inputs and outputs. Therefore the circularity score is 0.

Assumptions & free parameters 2 free parameters · 12 assumptions · 0 invented entities

The paper introduces no new physical entities. It pulls a large set of technical assumptions from the high-dimensional inference and sparse estimation literature; the most fragile are A8-A10, which are tailored to make the proof close.

free parameters (2)

lambda_u, lambda_v (sparse SVD tuning parameters)
Selected in practice by sample splitting and validation loss L(λ); the theory in A8 only requires them to lie in certain rate regimes, not specific values.
K (rank under test)
In the sequential procedure, K is determined from the data by the same test; in the theorems it is treated as known. The validity of the tests depends on K being small relative to the covariance spectrum.

assumptions (12)

domain assumption Z has independent subgaussian entries with fourth moment 3+Δ and product moment conditions up to order 8.
Needed for CLT of the U-statistics; excludes heavy-tailed and dependent noise.
domain assumption The Jacobian matrix B(µ0) from the Cragg-Donald reparameterization has smallest singular value bounded below.
Used only for the minimum discrepancy test result in Theorem 2.
domain assumption Elements of Π20 and V0 are uniformly bounded.
Regularity to avoid extreme values in the fixed-rank decomposition.
domain assumption pq = O(n^4), tr(Σ4) = o(tr^2(Σ2)), and ||Σ||_2^2 = o(tr(Σ2)).
Standard high-dimensional conditions ensuring the U-statistic variance is dominated by tr(Σ2).
domain assumption Singular values of Π0 are bounded away from zero and infinity.
Needed for identifiability and convergence of the sparse SVD estimator.
domain assumption The submatrix of Σ1 ⊗ Σ2 corresponding to the sparse support has bounded spectral norm.
Controls the effect of heteroscedastic errors on the restricted estimation problem.
domain assumption Z entries are subgaussian with uniform norm K0.
Tail bounds for the sparse SVD estimation error.
ad hoc to paper Tuning parameters λu, λv are chosen in specific rate regimes with concave penalties and negligible local concavity.
Conditions enough to make the oracle inequality for the penalized SVD estimator go through; no practical guidance for selecting these rates.
ad hoc to paper With probability tending to 1, the estimated supports contain the true nonzero rows or columns and have size κ_n = o(n).
States that the sparse SVD estimator recovers the support, which is part of what the theorem aims to prove and is not guaranteed by the algorithm.
ad hoc to paper ||\hat U\hat U^T - U0U0^T||_F = o_P(sqrt(tr(Σ2))/n) and the same for V.
This is the key link between sparse SVD convergence and the test's asymptotic normality; Theorem 4's rate satisfies it only if (k+l)n = o(tr(Σ2)), a sparsity condition not stated in Theorem 5.
domain assumption The mean matrix Π0 has nonzero entries only on k rows and l columns, with k,l small.
Sparse SVD theory and the plug-in statistic rely on this structural assumption.
ad hoc to paper The sparse SVD estimator is computed for the K used in the hypothesis test, and the test statistic for that K is calibrated as if K were fixed.
In the sequential procedure K is data-dependent; the theory does not account for this selection.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Hypothesis Testing for High-Dimensional Matrix-Valued Data." pith.science (2026). https://pith.science/paper/ESDTVD4N

@misc{pith2026241207987,
  author       = {Pith},
  title        = {Pith review of: Hypothesis Testing for High-Dimensional Matrix-Valued Data},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ESDTVD4N}},
  note         = {Machine review of arXiv:2412.07987}
}

read the original abstract

This paper addresses hypothesis testing for the mean of matrix-valued data in high-dimensional settings. We investigate the minimum discrepancy test, originally proposed by Cragg (1997), which serves as a rank test for lower-dimensional matrices. We evaluate the performance of this test as the matrix dimensions increase proportionally with the sample size, and identify its limitations when matrix dimensions significantly exceed the sample size. To address these challenges, we propose a new test statistic tailored for high-dimensional matrix rank testing. The oracle version of this statistic is analyzed to highlight its theoretical properties. Additionally, we develop a novel approach for constructing a sparse singular value decomposition (SVD) estimator for singular vectors, providing a comprehensive examination of its theoretical aspects. Using the sparse SVD estimator, we explore the properties of the sample version of our proposed statistic. The paper concludes with simulation studies and two case studies involving surveillance video data, demonstrating the practical utility of our proposed methods.

Figures

Figures reproduced from arXiv: 2412.07987 by the authors.

**Figure 2.** Left: matrix rank detected. Middle: matrix rank greater than 0 or not. Right: [PITH_FULL_IMAGE:figures/full_fig_p020_2.png] view at source ↗

**Figure 3.** Matrix rank detected 20 [PITH_FULL_IMAGE:figures/full_fig_p020_3.png] view at source ↗

Figures from the paper (2 more)

**Figure 4.** Figure 4: Sample video frames at different times. Left: background. Middle: one car. [PITH_FULL_IMAGE:figures/full_fig_p021_4.png]

**Figure 5.** Figure 5: Left: matrix rank detected. Right: number of cars in the parking lot (Ground [PITH_FULL_IMAGE:figures/full_fig_p022_5.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

A Short Note of Comparison between Convex and Non-convex Penalized Likelihood
stat.ME 2025-02 conditional novelty 2.0 of 10

Lasso is convex and fast but biased; SCAD and MCP reduce bias but are non-convex and slower, so method choice depends on context.

Reference graph

Works this paper leans on

16 extracted references · 16 canonical work pages · cited by 1 Pith paper

[1]

& Saranadasa, H

Bai, Z. & Saranadasa, H. (1996), ‘Effect of high dimension: by an example of a two sample problem’, Statistica Sinica6, 311–329. 22

work page 1996
[2]

& Xia, Y

Cai, T., Liu, W. & Xia, Y. (2014), ‘Two-sample test of high dimensional means under dependence’, Journal of the Royal Statistical Society: Series B (Statistical Methodology) 76(2), 349–372

work page 2014
[3]

X., Zhang, L.-X

Chen, S. X., Zhang, L.-X. & Zhong, P.-S. (2010), ‘Tests for high-dimensional covariance matrices’, Journal of the American Statistical Association105(490), 810–819

work page 2010
[4]

Cragg, J. G. & Donald, S. G. (1993), ‘Testing identifiability and specification in instrumen- tal variable models’, Econometric Theory9(2), 222–240

work page 1993
[5]

Cragg, J. G. & Donald, S. G. (1997), ‘Inferring the rank of a matrix’, Journal of Econo- metrics 76(1-2), 223–250

work page 1997
[6]

& Young, G

Eckart, C. & Young, G. (1936), ‘The approximation of one matrix by another of lower rank’, Psychometrika 1(3), 211–218

work page 1936
[7]

(1931), ‘The generalization of student’s ratio’, The Annals of Mathematical Statistics 2(3), 360–378

Hotelling, H. (1931), ‘The generalization of student’s ratio’, The Annals of Mathematical Statistics 2(3), 360–378

work page 1931
[8]

& Yang, S

Huang, Y., Li, C., Li, R. & Yang, S. (2022), ‘An overview of tests on high-dimensional means’, Journal of Multivariate Analysis188, 104813

work page 2022

Show all 16 references

[9]

Lee, M., Shen, H., Huang, J. Z. & Marron, J. (2010), ‘Biclustering via sparse singular value decomposition’, Biometrics 66(4), 1087–1095

2010
[10]

Li, D., Xue, L., Yang, H. & Yu, X. (2024), ‘Power-enhanced two-sample mean tests for high- dimensional compositional data with application to microbiome data analysis’, arXiv preprint arXiv:2405.02551

2024 arXiv
[11]

& Chen, S

Li, J. & Chen, S. X. (2012), ‘Two sample tests for high-dimensional covariance matrices’, The Annals of Statistics40(2), 908–940

2012
[12]

& Smith, R

Robin, J.-M. & Smith, R. J. (2000), ‘Tests of rank’, Econometric Theory16(2), 151–175. 23

2000
[13]

& Ishwar, P

Wang, Y., Jodoin, P.-M., Porikli, F., Konrad, J., Benezeth, Y. & Ishwar, P. (2014), Cdnet 2014: An expanded change detection benchmark dataset, in ‘Proceedings of the IEEE conference on computer vision and pattern recognition workshops’, pp. 387–394

2014
[14]

& Buja, A

Yang, D., Ma, Z. & Buja, A. (2016), ‘Rate optimal denoising of simultaneously sparse and low rank matrices’, Journal of Machine Learning Research17(1), 3163–3189

2016
[15]

Yu, X., Li, D., Xue, L. & Li, R. (2023), ‘Power-enhanced simultaneous test of high- dimensional mean vectors and covariance matrices with application to gene-set testing’, Journal of the American Statistical Association118(544), 2548–2561

2023
[16]

& Han, R

Zhang, A. & Han, R. (2019), ‘Optimal sparse singular value decomposition for high-dimensional high-order data’, Journal of the American Statistical Association 114(528), 1708–1725. 24

2019

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.

[1] [1]

& Saranadasa, H

Bai, Z. & Saranadasa, H. (1996), ‘Effect of high dimension: by an example of a two sample problem’, Statistica Sinica6, 311–329. 22

work page 1996

[2] [2]

& Xia, Y

Cai, T., Liu, W. & Xia, Y. (2014), ‘Two-sample test of high dimensional means under dependence’, Journal of the Royal Statistical Society: Series B (Statistical Methodology) 76(2), 349–372

work page 2014

[3] [3]

X., Zhang, L.-X

Chen, S. X., Zhang, L.-X. & Zhong, P.-S. (2010), ‘Tests for high-dimensional covariance matrices’, Journal of the American Statistical Association105(490), 810–819

work page 2010

[4] [4]

Cragg, J. G. & Donald, S. G. (1993), ‘Testing identifiability and specification in instrumen- tal variable models’, Econometric Theory9(2), 222–240

work page 1993

[5] [5]

Cragg, J. G. & Donald, S. G. (1997), ‘Inferring the rank of a matrix’, Journal of Econo- metrics 76(1-2), 223–250

work page 1997

[6] [6]

& Young, G

Eckart, C. & Young, G. (1936), ‘The approximation of one matrix by another of lower rank’, Psychometrika 1(3), 211–218

work page 1936

[7] [7]

(1931), ‘The generalization of student’s ratio’, The Annals of Mathematical Statistics 2(3), 360–378

Hotelling, H. (1931), ‘The generalization of student’s ratio’, The Annals of Mathematical Statistics 2(3), 360–378

work page 1931

[8] [8]

& Yang, S

Huang, Y., Li, C., Li, R. & Yang, S. (2022), ‘An overview of tests on high-dimensional means’, Journal of Multivariate Analysis188, 104813

work page 2022