Pith. sign in

REVIEW 2 major objections 5 minor 83 references

Optimal estimation of functionals of high-dimensional mean and covariance matrix

T0 review · 2 major / 5 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read Estimating μᵀΣ⁻¹μ has a sharp minimax rate, and the paper constructs an estimator that achieves it.

desk verdict The exact-sparse minimax results for μᵀΣ⁻¹μ are solid and worth serious refereeing, but the approximate-sparsity generalization in §3.4 is internally inconsistent and needs a major fix before publication. read the letter →

arxiv 1908.07460 v2 pith:SQGHEIYA submitted 2019-08-20 math.ST stat.TH

Jianqing Fan , Haolei Weng , Yifeng Zhou This is my paper · ORCID
classification math.STstat.TH MSC 62H1262C2062F12
keywords functionalestimationhigh-dimensionalstatisticsquadraticmeanandcovarianceminimaxoptimalityphasetransitionsparsitysub-gaussiandistribution
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

The paper asks how well one can estimate the single number $\theta = \mu^T \Sigma^{-1} \mu$ — the squared Sharpe ratio or the linear-discriminant signal — from $n$ high-dimensional samples when the vector $\alpha = \Sigma^{-1} \mu$ is sparse. It proves that without sparsity no consistent estimator exists when $p \ge n^2$, and that under sparsity the minimax absolute error is of order $[\tau \wedge (\tau+\sqrt{\tau})/\sqrt{n}] + [\tau \wedge (1+\tau) s \log p / n]$, up to constants. A bias-corrected plug-in estimator defined through an $\ell_1$-regularized estimate of $\alpha$ attains this rate whenever the signal $\tau$ is above $s \log p/n$; below that threshold the trivial estimator $0$ is already minimax optimal. Naive plug-in estimators with the same regularization fail, because the $\ell_1$ bias contaminates the quadratic term. The result matters because $\theta$ governs portfolio performance and classification difficulty, and knowing the exact rate tells practitioners when sparse estimation helps and when it cannot.

What carries the argument

The central object is the bias-corrected plug-in estimator $\tilde{\theta} = 2\hat{\mu}^T \tilde{\alpha} - \tilde{\alpha}^T \hat{\Sigma} \tilde{\alpha}$, built from the $\ell_1$-regularized estimator $\tilde{\alpha}$ defined by minimizing $\frac12 \beta^T\hat{\Sigma}\beta - \beta^T\hat{\mu} + \lambda\|\beta\|_1$ over a bounded ball. The estimator works because the KKT conditions for $\tilde{\alpha}$ express its main bias as $-\lambda \Sigma^{-1} \hat{g}$, and replacing $\Sigma^{-1}\hat{\mu}$ by $\tilde{\alpha}$ cancels that bias in the plug-in product. The proofs show that on a high-probability event the error $\tilde{\alpha}-\alpha$ lies in a sparse cone where coordinates off the support are controlled by three times the on-support coordinates, which yields the $\ell_2$ bound. Lower bounds come from two-point and fuzzy-hypothesis tests with $\chi^2$-divergence computations for the matching minimax statements.

What would settle it

Run the debiased estimator and the best $c\ne 2$ plug-in on Gaussian data with $s$ equal nonzero entries in $\alpha$, $\tau$ constant, and $\tau \gg s\log p/n$; if the plug-in's worst-case error falls below order $\sqrt{\tau(1+\tau)}\, s\log p/n$, or the debiased error exceeds order $(\tau+\sqrt{\tau})/\sqrt{n} + (1+\tau)s\log p/n$, the rate statements are wrong.

Watch

Extended reading notes

Core claim

The central discovery is that the estimation error for $\theta$ undergoes a sharp phase transition between a parametric rate and a high-dimensional sparse rate. On the class $H(s,\tau)$ where $\alpha=\Sigma^{-1}\mu$ has at most $s$ nonzero entries and $\theta\le\tau$, with eigenvalues of $\Sigma$ between fixed constants, the minimax rate is $[\tau \wedge (\tau+\sqrt{\tau})/\sqrt{n}] + [\tau \wedge (1+\tau)s \log p / n]$. The bias-corrected estimator $\tilde{\theta} = 2\hat{\mu}^T \tilde{\alpha} - \tilde{\alpha}^T \hat{\Sigma} \tilde{\alpha}$, computed from an $\ell_1$-regularized M-estimator $\tilde{\alpha}$ of $\alpha$, achieves this rate and is minimax optimal when $\tau \gtrsim s\log p/n$; when $\tau \lesssim s\log p/n$ the trivial estimator $0$ is optimal. A family of plug-in estimators $\tilde{\theta}_c = c\hat{\mu}^T\tilde{\alpha} + (1-c)\tilde{\alpha}^T\hat{\Sigma}\tilde{\alpha}$ is suboptimal for every constant $c\ne 2$, with error at least of order $\sqrt{\tau(1+\tau)}\, s\log p / n$. In the dense regime $s\gtrsim\sqrt{p}$, a different de-biasing — an unbiased inverse-Wishart correction for Gaussian data and an iterated data-splitting bias correction for sub-Gaussian data — gives rate $\sqrt{p}/n + (\tau+\sqrt{\tau})/\sqrt{n}$, completing an elbow at $s\approx\sqrt{p}$.

Load-bearing premise

Everything hinges on $\Sigma^{-1}\mu$ being sparse (or nearly so) while the eigenvalues of $\Sigma$ stay between two fixed constants; without that, Proposition 1 shows the functional cannot be consistently estimated when $p \ge n^2$.

Editorial extensions

If this is right

  • When the signal is strong enough ($\tau \gtrsim s\log p/n$), the debiased estimator achieves the minimax rate, so no other estimator can do uniformly better over the sparse class.
  • When $\tau \lesssim s\log p/n$, estimating by zero is optimal; this is a genuine regime where the signal is too weak for any data-based estimator to beat the trivial guess.
  • Naive plug-in estimators built from the same $\ell_1$ estimate are provably suboptimal for every constant $c\ne 2$, so the de-biasing step is rate-determining, not cosmetic.
  • In the dense regime $s\gtrsim\sqrt{p}$, the minimax rate becomes $(1+\tau)\sqrt{p}/n + (\tau+\sqrt{\tau})/\sqrt{n}$, giving an elbow at $s\approx\sqrt{p}$.
  • The same de-biased scheme with robust estimators of mean and covariance extends the rates (with an extra factor $s$ for estimating $\alpha$) to heavy-tailed distributions with bounded fourth moments.

Reading between the lines

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

  • The paper's phase-transition analysis implies that reported Sharpe ratios or LDA signal estimates in high dimension should come with a regime flag: below $\tau \approx s\log p/n$, the estimate is essentially indistinguishable from zero, and confidence statements need to reflect that.
  • Because the suboptimality of $c\ne 2$ plug-ins is driven by $\lambda\|\tilde{\alpha}\|_1$, any feasible estimator of $\theta$ built from a sparse $\alpha$ must either remove this bias or use an $\ell_0$-type construction; a testable extension is to check whether Dantzig-selector plug-ins obey the same gap outside the paper's theory.
  • The open scaling $p\lesssim s^2$ with $n\lesssim p\lesssim n^2$ suggests there should be an interpolation between the sparse and dense rates; a simulation study sweeping $(n,p,s,\tau)$ across that boundary could reveal the missing minimax formula.
  • Since $\theta$ controls the Bayes error in linear discriminant analysis, the minimax gap here quantifies precisely when sparse LDA can be expected to beat random guessing in high dimensions.
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.

Referee Report

2 major / 5 minor

Summary. The manuscript studies minimax estimation of the quadratic functional θ = μ^T Σ^{-1} μ from n i.i.d. sub-Gaussian vectors in R^p, under the assumption that α = Σ^{-1} μ is exactly or approximately sparse and that the eigenvalues of Σ are bounded. It proposes a debiased ℓ1-regularized plug-in estimator (3.1)–(3.2), proves upper bounds (Theorem 1), matching lower bounds (Theorem 2), and establishes a phase transition at τ ≍ s log p/n (Corollary 1). It further shows that a family of naive plug-in estimators with c ≠ 2 is suboptimal (Proposition 2), studies an ℓ0 variant (Theorem 3), extends the framework to robust inputs of mean and covariance estimators (Section 3.3), to approximate sparsity classes (Section 3.4), and to the dense regime (Section 3.5), and reports simulation and S&P500 portfolio experiments (Section 4).

Significance. If the results stood as stated, the paper would make a substantial contribution to high-dimensional functional estimation with unknown covariance, extending Fan, Rigollet, and Wang (2015) and Collier, Comminges, and Tsybakov (2017) to the setting where Σ is unknown and α = Σ^{-1} μ is sparse. The strengths are the complete proofs in Section 5, the explicit two-point and χ² lower-bound constructions, the feasible debiased estimator that attains the exact-sparse rate, and the detailed treatment of the dense regime. However, the approximate-sparsity generalization in Section 3.4 is internally inconsistent with the exact-sparse results when q = 0, and the exponential factor in Theorem 2(b) prevents the claimed Corollary 1(b) lower bound as written. These defects are localized and likely reparable, but they affect two advertised central claims, so the paper needs a major revision before acceptance.

major comments (2)
  1. [Theorem 2(b) and the derivation of Corollary 1(b)] The claim that setting q = 0 in the approximate-sparsity results recovers the exact-sparse results is false with the displayed formulas. Since H_0(R, τ) = H(R, τ) under the paper's 0^0 = 0 convention, Corollary 2(b) with q = 0 must coincide with Corollary 1(b). Instead, q = 0 in Corollary 2(b) yields (1 + τ)^{1/2} R (log p/n)^{1/2} with a phase transition at τ ≍ R (log p/n)^{1/2}, whereas Corollary 1(b) yields [τ ∧ (1 + τ) s log p/n] + [τ ∧ (τ + √τ)/√n] with a transition at τ ≍ s log p/n. The proof in Section 5.7.1 shows which rate the lower-bound construction actually supports: in case (i) the proof substitutes the effective sparsity s_eff = R̃ (log p/n)^{-q/2} τ̃^{-q/2} into the exact-sparse χ² construction and obtains a rate of order τ̃^{1-q/2} R̃ (log p/n)^{1-q/2}, consistent with exponents 1 - q/2, not (1 - q)/2. Thus the displayed exponents in Theorem 4, Theorem 5, and Corollary 2 need to be corrected, and the sentence 'Setting q = 0 in the two theorems, we fully recover (3.7)–(3.10)' must be revised accordingly. The scaling conditions in Theorem 4 and Corollary 2 should also be rechecked after this correction.
  2. [Theorem 2(b) and the derivation of Corollary 1(b)] The lower bound in Theorem 2(b) contains the factor c_0 exp(-e^{2s^2 p^{c_6 c_0^{-1}}}) with c_0 ∈ [0,1] and c_6 > 0. For p ≥ 2 and s ≥ 1, this factor is exponentially small for every admissible c_0, and the argument after (3.10) that one can 'choose sufficiently small c_0' to make e^{2s^2 p^{c_6 c_0^{-1}} = 1 + o(1) is not valid; decreasing c_0 increases the positive exponent c_6 c_0^{-1}. As stated, the second displayed term in Theorem 2(b) therefore does not yield the claimed [τ ∧ (1 + τ) s log p/n] lower bound in Corollary 1(b). The authors need to restate the exponential factor (for example, with a negative exponent on p, if that is the intended expression) and the precise condition under which the factor is bounded below by a universal constant.
minor comments (5)
  1. [Section 4.2] The phrase 'Shape ratio' in the discussion of Table 1 should be 'Sharpe ratio'.
  2. [Section 3.1, Theorem 1] The reuse of the symbol c both as an arbitrary positive constant in the exponent and as the constant controlling the scaling condition involving c̃ is confusing; please use distinct symbols.
  3. [Section 3.3, Proposition 3] The condition (3.17) involves an unspecified sample size m; in the applications to the one-sample and two-sample problems, please clarify whether m is n, n1, or n2 and how it enters the subsequent displayed bounds.
  4. [Section 4.1, Figure 1] The theoretical rate functions f_α, g_α, h_α and f_θ, g_θ, h_θ are plotted with calibrated constants, but the text does not explicitly state which theorem or corollary each dashed curve corresponds to; adding this mapping would improve reproducibility.
  5. [Section 5.3.1] In the proof of Theorem 2(b), the chi-squared calculations are stated to follow from Lemma A.1 of Fan, Rigollet, and Wang (2015); since this lemma is central to the lower bound, stating it explicitly in Section 5.9 would make the proof more self-contained.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the minimax lower and upper bounds are derived from explicit data-generating constructions and concentration inequalities, with self-citations only to peripheral technical lemmas.

full rationale

The derivation chain is self-contained. The upper bounds (Theorems 1 and 4) follow from KKT identities, the cone/restricted-eigenvalue lemmas (Lemmas 3–4), and standard concentration inequalities (Lemmas 1–2); the constants are selected from the parameter-class bounds, not from the target error rates. The lower bounds (Theorems 2 and 5) are obtained from explicit two-point and two-fuzzy-hypothesis Gaussian constructions with computed KL/χ² divergences, independently of the estimator's form; Proposition 1 is a by-product of these constructions. The only self-citation that enters the proof, Fan, Rigollet, and Wang (2015), supplies a generic algebraic bound on χ² divergence for Gaussian mixtures (their Lemma A.1); it does not contain the target functional's rate and is parameter-free with stated assumptions, so it is genuine external support rather than a load-bearing self-citation. The robust-input results cited from Ke et al. and the de-biasing references (Zhang–Zhang, Javanmard–Montanari) are similarly auxiliary. No fitted parameter is renamed as a prediction, and no quantity is defined in terms of the target. The apparent failure of the displayed approximate-sparsity rates to reduce to Corollary 1 when q = 0 is an internal consistency and correctness concern in Section 3.4, not a circularity: it does not make the claimed rates equivalent to the paper's inputs by construction.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

The central claim rests on the sub-gaussian and sparsity assumptions, which are explicitly stated. No new objects beyond the debiased estimator are introduced. The free parameter entry reflects a limitation in the simulation section only.

free parameters (1)
  • Simulation tuning parameters (λ, γ) in Section 4.1 = unknown, chosen to minimize estimation error
    In the first set of simulations, tuning parameters are picked optimally to minimize the estimation error of α, which uses oracle information and may make the empirical rates look better than a fully data-driven procedure would. This does not affect the theoretical claims but limits the reproducibility of the figures.
assumptions (4)
  • domain assumption Sub-gaussianity of x = Σ^{1/2} y + μ with ||y||_{ψ2} = ν as in (2.1).
    Assumed throughout Section 3, used for concentration inequalities in Lemmas 1 and 2.
  • domain assumption Eigenvalue bounds c_L ≤ λ_min(Σ) ≤ δ_Σ ≤ c_U for fixed constants c_L, c_U > 0.
    Part of the parameter spaces H(s,τ) and H_q(R,τ), used to control quadratic forms and ensure identifiability.
  • domain assumption Sparsity of α = Σ^{-1} μ: exact (H(s,τ)) or approximate (H_q(R,τ)).
    The entire analysis rests on this structural assumption; Proposition 1 shows the functional is not estimable in general without it.
  • standard math Standard concentration inequalities (Hoeffding, Bernstein, matrix deviation) from Vershynin (2018) and related references.
    Used in the proofs of Lemmas 1, 2 and the main theorems; these are unproved background results.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Optimal estimation of functionals of high-dimensional mean and covariance matrix." pith.science (2026). https://pith.science/paper/SQGHEIYA

@misc{pith2026190807460,
  author       = {Pith},
  title        = {Pith review of: Optimal estimation of functionals of high-dimensional mean and covariance matrix},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/SQGHEIYA}},
  note         = {Machine review of arXiv:1908.07460}
}
abstract

Motivated by portfolio allocation and linear discriminant analysis, we consider estimating a functional $\mathbf{\mu}^T \mathbf{\Sigma}^{-1} \mathbf{\mu}$ involving both the mean vector $\mathbf{\mu}$ and covariance matrix $\mathbf{\Sigma}$. We study the minimax estimation of the functional in the high-dimensional setting where $\mathbf{\Sigma}^{-1} \mathbf{\mu}$ is sparse. Akin to past works on functional estimation, we show that the optimal rate for estimating the functional undergoes a phase transition between regular parametric rate and some form of high-dimensional estimation rate. We further show that the optimal rate is attained by a carefully designed plug-in estimator based on de-biasing, while a family of naive plug-in estimators are proved to fall short. We further generalize the estimation problem and techniques that allow robust inputs of mean and covariance matrix estimators. Extensive numerical experiments lend further supports to our theoretical results.

Figures

Figures reproduced from arXiv: 1908.07460 by the authors.

Figure 1
Figure 1. Averaged error v.s. sample size on logarithmic scale for the three settings (solid curves) [PITH_FULL_IMAGE:figures/full_fig_p027_1.png] view at source ↗
Figure 2
Figure 2. Comparison of different optimally tuned estimators for [PITH_FULL_IMAGE:figures/full_fig_p030_2.png] view at source ↗
Figure 3
Figure 3. Comparison of different cross-validation tuned estimators for [PITH_FULL_IMAGE:figures/full_fig_p030_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Comparison of different optimally tuned estimators for [PITH_FULL_IMAGE:figures/full_fig_p031_4.png]
Figure 5
Figure 5. Figure 5: Comparison of different cross-validation tuned estimators for [PITH_FULL_IMAGE:figures/full_fig_p032_5.png]
Figure 6
Figure 6. Figure 6: Portfolio performance. market, and the same phenomenon can be observed for the minimum variance portfolio with gross exposure constraint, which is expected to performance in terms of stability and maximum draw￾down. Value-Weighted Equal-Weighted Gross Exposure Sparse P…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

83 extracted references · 77 canonical work pages

  1. [1]

    Anderson, T. (2003). An Introduction to Multivariate Statistical Analysis, 3rd Edition . Wiley- Interscience

    2003

  2. [2]

    Ao, M., Li, Y., and Zheng, X. (2017). Solving the Markowitz Optimization Problem for Large Portfolios. Manuscript

    2017

  3. [3]

    Bai, Z., Liu, H., and Wong, W. (2009). Enhancement of the applicability of Markowitz’s portfolio optimization by utilizing random matrix theory. Mathematical Finance, 19(4), 639–667

    2009

  4. [4]

    and Sarandasa, H

    Bai, Z. and Sarandasa, H. (1996). Effect of High Dimension: By an Example of a Two Sample Problem. Statistica Sinica, 6, 311–329

    work page 1996

  5. [5]

    Bai, Z., Silverstein, J., and Yin, Y. (1988). A note on the largest eigenvalue of a large dimensional sample covariance matrix. Journal of Multivariate Analysis , 26, 166–168. 83

    work page 1988

  6. [6]

    and Yin, Y

    Bai, Z. and Yin, Y. (1993). Limit of the smallest eigenvalue of large dimensional sample covariance matrix. Annals of Probability, 21, 1275–1294

    work page 1993

  7. [7]

    and Levina, E

    Bickel, P. and Levina, E. (2004). Some theory for Fisher’s linear discriminant function, ‘naive Bayes’, and some alternatives when there are many more variables than observations. Bernoulli, 10, 989–1010

    work page 2004

  8. [8]

    and Ritov, Y

    Bickel, P. and Ritov, Y. (1988). Estimating integrated squared density derivatives: Sharp best order of convergence estimates. Sankhy¯ a Ser. A, 50, 381–393

    work page 1988

Show all 83 references
  1. [9]

    Bickel, P., Ritov, Y., and Tsybakov, A. (2009). Simultaneous analysis of Lasso and Dantzig selector. Annals of Statistics , 37(4), 1705–1732. Birg´ e, L. and Massart, P. (2001). Gaussian model selection.Journal of the European Mathematical Society, 3(3), 203–268

    2009

  2. [10]

    Boucheron, S., Lugosi, G., and Massart, P. (2013). Concentration inequalities: A nonasymptotic theory of independence. Oxford University Press

    2013

  3. [11]

    Brodie, J., Daubechies, I., De Mol, C., Giannone, D., and Loris, I. (2009). Sparse and stable Markowitz portfolios. Proceedings of the National Academy of Sciences , 106(30), 12267–12272. B¨ uhlmann, P. and van de Geer, S. (2011). Statistics for high-dimensional data: methods,...

    2009

  4. [12]

    Butucea, C. (2007). Goodness-of-fit testing and quadratic functional estimation from indirect ob- servations. Annals of Statistics , 35, 1907–1930

    2007

  5. [13]

    and Comte, F

    Butucea, C. and Comte, F. (2009). Adaptive estimation of linear functionals in the convolution model and applications. Bernoulli, 15, 69–98

    2009

  6. [14]

    and Guo, Z

    Cai, T. and Guo, Z. (2017). Confidence intervals for high-dimensional linear regression: Minimax rates and adaptivity. Annals of Statistics , 45(2), 615–646

    2017

  7. [15]

    Cai, T., Liu, W., and Xia, Y. (2014). Two-sample test of high dimensional means under dependence. J.R.Statist. Soc. B , 76, 349–372

    2014

  8. [16]

    and Low, M

    Cai, T. and Low, M. (2004). Minimax estimation of linear functionals over nonconvex parameter spaces. Annals of Statistics , 32(2), 552–576. 84

    2004

  9. [17]

    Cai, T., Ren, Z., and Zhou, H. (2016). Estimating structured high-dimensional covariance and precision matrices: Optimal rates and adaptive estimation. 10, 1–59

    2016

  10. [18]

    and Yuan, M

    Cai, T. and Yuan, M. (2012). Adaptive covariance matrix estimation through block thresholding. Annals of Statistics , 40, 2014–2042

    2012

  11. [19]

    Cai, T., Zhang, C., and Zhou, H. (2010). Optimal rates of convergence for covariance matrix estimation. Annals of Statistics , 38(4), 2118–2144

    2010

  12. [20]

    and Zhang, L

    Cai, T. and Zhang, L. (2018). High-dimensional Linear Discriminant Analysis: Optimality, Adap- tive Algorithm, and Missing Data. Manuscript

    2018

  13. [21]

    and Zhou, H

    Cai, T. and Zhou, H. (2012). Optimal rates of convergence for sparse covariance matrix estimation. Annals of Statistics , 40(5), 2389–2420. Cand` es, E. and Tao, T. (2007). The Dantzig selector: Statistical estimation whenp is much larger than n. Annals of Statistics , 35(6), ...

    2012

  14. [22]

    and Qin, Y

    Chen, S. and Qin, Y. (2010). A Two-sample Test for High-dimensional Data with Application to Gene-Set Testing. Annals of Statistics , 38, 808–835

    2010

  15. [23]

    Chen, X., Xu, M., and Wu, W. (2016). Regularized estimation of linear functionals of precision matrices for high-dimensional time series. IEEE Transactions on Signal Processing , 64(24), 6459–6470

    2016

  16. [24]

    Collier, O., Comminges, L., and Tsybakov, A. (2017). Minimax estimation of linear and quadratic functionals on sparsity classes. Annals of Statistics , 45(3), 923–958

    2017

  17. [25]

    and Thomas, J

    Cover, T. and Thomas, J. (2012). Elements of information theory . John Wiley & Sons

    2012

  18. [26]

    DeMiguel, V., Garlappi, L., Nogales, J., Uppal, R. (2009). A generalized approach to portfolio optimization: Improving performance by constraining portfolio norms. Management Science , 55(5), 798–812

    2009

  19. [27]

    Devroye, L., Lerasle, M., Lugosi, G., and Oliveira, R. (2016). Sub-Gaussian mean estimators. Annals of Statistics , 44(6), 2695–2725

    2016

  20. [28]

    and Nussbaum, M

    Donoho, D. and Nussbaum, M. (1990). Minimax quadratic estimation of a quadratic functional. J. Complexity, 6, 290–323. 85

    1990

  21. [29]

    and Low, M

    Efromovich, S. and Low, M. (1996). On optimal adaptive estimation of a quadratic functional. Annals of Statistics , 24, 1106–1125

    1996

  22. [30]

    Fan, J. (1991). On the estimation of quadratic functionals. Annals of Statistics , 19(3), 1273–1294

    1991

  23. [31]

    and Fan, Y

    Fan, J. and Fan, Y. (2008). High dimensional classification using features annealed independence rules. Annals of Statistics , 36, 2605–2637

    2008

  24. [32]

    Fan, J., Fan, Y., and Lv, J. (2008). High dimensional covariance matrix estimation using a factor model. Journal of Econometrics , 147, 186–197

    2008

  25. [33]

    Fan, J., Han, F., and Liu, H. (2014). Challenges of big data analysis. National science review, 1(2), 293–314

    2014

  26. [34]

    and Li, R

    Fan, J. and Li, R. (2001). Variable selection via nonconcave penalized likelihood and its oracle properties. Journal of the American Statistical Association , 96, 1348–1360

    2001

  27. [35]

    Fan, J., Liao, Y., and Liu, H. (2016). An overview of the estimation of large covariance and precision matrices. Econometrics Journal, 19, 1–32

    2016

  28. [36]

    Fan, J., Liao, Y., and Mincheva, M. (2013). High-dimensional covariance matrix estimation in approximate factor models. Annals of Statistics , 39, 3320–3356

    2013

  29. [37]

    Fan, J., Liao, Y., and Mincheva, M. (2013). Large covariance estimation by thresholding principal orthogonal complements. J. R. Statisti. Soc. B , 75, 603–680

    2013

  30. [38]

    and Lv, J

    Fan, J. and Lv, J. (2008). Sure independence screening for ultrahigh dimensional feature space. Journal of the Royal Statistical Society: Series B (Statistical Methodology) , 70(5), 849–911

    2008

  31. [39]

    and Lv, J

    Fan, J. and Lv, J. (2018). Sure independence screening.Wiley StatsRef: Statistics Reference Online, 1–8

    2018

  32. [40]

    Fan, J., Rigollet, P., and Wang, W. (2015). Estimation of functionals of sparse covariance matrices. Annals of Statistics , 43(6), 2706–2737

    2015

  33. [41]

    Fan, J., Wang, K., Zhong, Y., and Zhu, Z. (2018). Robust high dimensional factor models with applications to statistical machine learning. Statistical Science, invited

    2018

  34. [42]

    Fan, J., Zhang, J., and Yu, K. (2012). Vast portfolio selection with gross exposure constraints. Journal of the American Statistical Association , 107(498), 592–606

    2012

  35. [43]

    Fan, J., Wang, W., and Zhu, Z.W. (2021+). A shrinkage principle for heavy-tailed data: High- dimensional robust low-rank matrix recovery. Annals of Statistics , to appear

    2021

  36. [44]

    and Bengtsson, T

    Furrer, R. and Bengtsson, T. (2007). . Estimation of high-dimensional prior and posterior covariance matrices in Kalman filter variants. Journal of Multivariate Analysis , 98(2), 227–255

    2007

  37. [45]

    Guo, Z., Wang, W., Cai, T., and Li, H. (2018). Optimal Estimation of Genetic Relatedness in High-Dimensional Linear Models. Journal of the American Statistical Association , 0, 1–12

    2018

  38. [46]

    and Khas’minskii, R

    Ibragimov, I. and Khas’minskii, R. (1984). Nonparametric estimation of the value of a linear func- tional in Gaussian white noise. Theory Probab. Appl., 29, 18–32. 86

    1984

  39. [47]

    and Montanari, A

    Javanmard, A. and Montanari, A. (2014). Confidence intervals and hypothesis testing for high dimensional regression. Journal of Machine Learning Research, , 15(1), 2869–2909

    2014

  40. [48]

    and Montanari, A

    Javanmard, A. and Montanari, A. (2014). Hypothesis Testing in High-Dimensional Regression under the Gaussian Random Design Model: Asymptotic Theory.IEEE Trans. on Inform. Theory, 60(10), 6522–6554

    2014

  41. [49]

    and Montanari, A

    Javanmard, A. and Montanari, A. (2018). Debiasing the lasso: Optimal sample size for Gaussian designs. Annals of Statistics , 46(6), 2593–2622

    2018

  42. [50]

    Johnstone, I. (2001). On the distribution of the largest eigenvalue in principal components analysis. Annals of Statistics , 29, 295–327

    2001

  43. [51]

    and Lu, Y

    Johnstone, I. and Lu, Y. (2009). On consistency and sparsity for principal components analysis in high dimensions. Journal of the American Statistical Association , 104(486), 682–693

    2009

  44. [52]

    and Titterington, M

    Johnstone, I. and Titterington, M. (2009). Statistical challenges of high-dimensional data. Phil. Trans. R. Soc. A, 364, 4237–4253

    2009

  45. [53]

    and Zhou, G

    Kan, R. and Zhou, G. (2007). Optimal portfolio choice with parameter uncertainty. Journal of Financial and Quantitative Analysis , 42(3), 621–656

    2007

  46. [54]

    Karoui, N. (2008). Operator norm consistent estimation of large-dimensional sparse covariance matrices. Annals of Statistics , 36(6), 2717–2756

    2008

  47. [55]

    Karoui, N. (2010). High-dimensionality effects in the Markowitz problem and other quadratic pro- grams with linear constraints: Risk underestimation. Annals of Statistics , 38(6), 3487–3566

    2010

  48. [56]

    Ke, Y., Minsker, S., Ren, Z., Sun, Q., and Zhou, W. (2019). User-Friendly Covariance Estimation for Heavy-Tailed Distributions. arXiv:1811.01520. Klemel¨ a, J. and Tsybakov, A. (2001). Sharp adaptive estimation of linear functionals. Annals of Statistics, 29, 1567–1600

    2019 arXiv

  49. [57]

    Koltchinskii, V. (2019). Asymptotically efficient estimation of smooth functionals of covariance operators. arXiv: 1710.09072

    2019 arXiv

  50. [58]

    Koltchinskii, V. (2020). Estimation of smooth functionals in high-dimensional models: bootstrap chains and Gaussian approximation. arXiv: 2011.03789

    2020 arXiv

  51. [59]

    and Zhilova, M

    Koltchinskii, V. and Zhilova, M. (2019). Estimation of smooth functionals in normal models: bias reduction and asymptotic efficiency. arXiv: 1912.08877

    2019 arXiv

  52. [60]

    and Fan, J

    Lam, C. and Fan, J. (2009). Sparsistency and rates of convergence in large covariance matrix estimation. Annals of Statistics , 37, 4254–4278

    2009

  53. [61]

    and Massart, P

    Laurent, B. and Massart, P. (2000). Adaptive estimation of a quadratic functional by model selec- tion. Annals of Statistics , 28, 1302–1338

    2000

  54. [62]

    and Wainwright, M

    Loh, P. and Wainwright, M. (2017). Support recovery without incoherence: A case for nonconvex regularization. Annals of Statistics , 45(6), 2455–2482. 87

    2017

  55. [63]

    Markowitz, H. (1952). Portfolio selection. Journal of Finance , 7, 77–91

    1952

  56. [64]

    Mai, Q. (2013). A review of discriminant analysis in high dimensions. Computational Statistics, 5, 190–197

    2013

  57. [65]

    A direct approach to sparse discriminant analysis in ultra- high dimensions

    Mai, Q., Zou, H., and Yuan, M.(2012). A direct approach to sparse discriminant analysis in ultra- high dimensions. Biometrika, 99, 29–42

    2012

  58. [66]

    Negahban, S., Ravikumar, P., Wainwright, M., and Yu, B. (2012). A unified framework for high- dimensional analysis of M-estimators with decomposable regularizers. Statistical Science, 27(4), 538–557

    2012

  59. [67]

    Nemirovskii, A. (2000). Topics in nonparametric statistics. Lectures on Probability Theory and Statistics (Saint-Flour, 1998). Lecture Notes in Math . 1738, 85–277. Springer, Berlin

    2000

  60. [68]

    Paul, D. (2007). Asymptotics of sample eigenstructure for a large dimensional spiked covariance model. Statistica Sinica, 17, 1617–1642

    2007

  61. [69]

    Raskutti, G., Wainwright, M., and Yu, B. (2010). Restricted eigenvalue properties for correlated Gaussian designs. Journal of Machine Learning Research , 11, 2241–2259

    2010

  62. [70]

    and Wang, S

    Shao, J., Wang, Y., Deng, X. and Wang, S. (2011). Sparse linear discriminant analysis with high dimensional data. Annals of Statistics , 39, 1241–1265

    2011

  63. [71]

    Srivastava, M. (2009). A test for the mean vector with fewer observations than the dimension under non-normality. Journal of Multivariate Analysis , 100, 518–532

    2009

  64. [72]

    and Du, M

    Srivastava, M. and Du, M. (2008). A Test for the Mean Vector with Fewer Observations than the Dimension. Journal of Multivariate Analysis , 99, 386–402

    2008

  65. [73]

    Sun, Q., Zhou, W., and Fan, J. (2018). Adaptive huber regression. Journal of the American Sta- tistical Association, 1–35

    2018

  66. [74]

    Tibshirani, R. (1996). Regression shrinkage and selection via the lasso. Journal of the Royal Sta- tistical Society. Series B , 58(1), 267–288

    1996

  67. [75]

    Tsybakov, A. (2009). Introduction to Nonparametric Estimation. Springer, New York. Van de Geer, S., B¨ uhlmann, P., Ritov, Y., and Dezeure, R. (2014). On asymptotically optimal confidence regions and tests for high-dimensional models. Annals of Statistics, 42(3), 1166–1202. Van...

    2009

  68. [76]

    Vershynin, R. (2010). Introduction to the non-asymptotic analysis of random matrices. arXiv:1011.3027

    2010 arXiv

  69. [77]

    Vershynin, R. (2018). High-dimensional probability: An introduction with applications in data sci- ence. Cambridge University Press. Von Rosen, D. (1988). Moments for the inverted Wishart distribution. Scandinavian Journal of Statistics, 15, 97–109. 88

    2018

  70. [78]

    Wainwright, M. (2019). High-dimensional statistics: a non-asymptotic viewpoint . Cambridge Uni- versity Press

    2019

  71. [79]

    Wang, L., Peng, B., and Li, R. (2015). A High-Dimensional Nonparametric Multivariate Test for Mean Vector. Journal of the American Statistical Association , 110, 1658–1669

    2015

  72. [80]

    and Fan, J

    Wang, W. and Fan, J. (2017). Asymptotics of empirical eigen-structure for high dimensional spiked covariance. Annals of Statistics , 45, 1342-1374

    2017

  73. [81]

    Zhang, C. (2010). Nearly unbiased variable selection under minimax concave penalty. The Annals of Statistics, 38, 894–942

    2010

  74. [82]

    and Zhang, S

    Zhang, C. and Zhang, S. (2014). Confidence intervals for low dimensional parameters in high dimen- sional linear models. Journal of the Royal Statistical Society: Series B (Statistical Methodology) , 76(1), 217–242

    2014

  75. [83]

    Zhong, P., Chen, S., and Xu, M. (2013). Tests alternative to higher criticism for high-dimensional means under sparsity and column-wise dependence. Annals of Statistics , 6, 2820–2851. 89

    2013

Pith tools

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