Bayesian optimal change point detection in high-dimensions

Jaehoon Kim; Kyoungjae Lee; Lizhen Lin

REVIEW 4 major objections 6 minor 5 references

Bayesian optimal change point detection in high-dimensions

T0 review · 4 major / 6 minor · reviewed 2026-08-12 · 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 the first Bayesian change point detection methods for high-dimensional mean and covariance structures, proving consistency under milder conditions and near-minimax localization rates.

desk verdict A genuine first Bayesian method for high-dimensional change point detection, but the implemented alpha-calibration step falls outside the theorem conditions, so the consistency claims are conditional. read the letter →

arxiv 2411.14864 v1 pith:73H5PI7K submitted 2024-11-22 stat.ME

Jaehoon Kim , Kyoungjae Lee , Lizhen Lin This is my paper · ORCID

classification stat.ME MSC 62F1562G2062H15

keywords high-dimensionalchangepointdetectionBayesianhypothesistestingmaximumpairwiseBayesfactormeanvectorcovariancematrixminimaxlocalizationratemultiscalemethodempiricalfalsepositive

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

Change points in high-dimensional time series—shifts in the mean vector or the covariance matrix—are hard to find because the dimension can exceed the sample size and the change may involve only a few coordinates. This paper proposes the first Bayesian procedures for this problem: it tests each candidate location and each coordinate (or coordinate pair) with a windowed Bayes factor and takes the maximum as evidence of a change. The paper proves that these tests are consistent under both null and alternative, that they estimate the number and locations of change points with error bounded by the chosen window size, and that this localization rate is minimax-optimal up to a logarithmic factor. The conditions are milder than those of existing frequentist methods: the number of changed coordinates need not grow with the dimension, and no sparsity of the covariance or precision matrix is required.

What carries the argument

The central object is the maximum pairwise Bayes factor (mxPBF): for each candidate location $l$ and window size $n_w$, compute a Bayes factor for a change in each coordinate (mean case) or each pair of coordinates conditional on one another (covariance case) using only the $n_w$ observations before and after $l$, then take the maximum over coordinates or pairs and over candidate locations. The priors are centered at local sample means or local regression coefficients with variance scaled by $\gamma_{n_w} = (n_w \vee p)^{-\alpha}$, which makes the log Bayes factors grow or shrink at polynomial rates in $n_w \vee p$. The theory combines tail bounds on these pairwise Bayes factors with union bounds over locations and coordinates; the multiscale variant groups detections from several window sizes by a majority vote, and the localization guarantee follows because the refined estimate is chosen as the argmax within the window after the first threshold crossing.

What would settle it

Simulate from the null hypothesis $X_i \sim N_p(0, I_p)$ with $n = 500$, $p = 1000$, window $n_w = 50$, threshold $C_{cp} = 10$, and $\alpha$ chosen by the paper's FPR calibration with $N = 300$; if the proportion of null datasets in which the mxPBF exceeds $C_{cp}$ grows with $p$ instead of staying near the calibrated 0.05, then the implemented version does not have the null consistency that Theorem 2.1 claims for the theoretical $\alpha$.

Watch

Extended reading notes

Core claim

The paper establishes that change points in the mean vector or covariance matrix of a high-dimensional Gaussian sequence can be detected and localized by a maximum pairwise Bayes factor computed on small windows around each candidate location. For the mean, Theorems 2.1 and 2.2 prove that the test is consistent and that the estimated number and locations satisfy $P_0(\hat K = K_0, \max_k |i_k - \hat i_k| \le n_w) \to 1$ whenever the minimum scaled componentwise signal exceeds roughly $\log(n_w \vee p)/n_w$ and certain window, sparsity, and hyperparameter conditions hold; Theorem 2.3 gives a matching minimax lower bound up to a logarithmic factor. For the covariance, Theorems 3.1 and 3.2 give analogous guarantees under signal conditions (A3) or (A3*), and Theorems 3.3 and 3.4 show that the localization rate $n_w/n$ is minimax-optimal up to $\log(n_w \vee p)$. The authors contend that this is the first Bayesian treatment of high-dimensional change point detection and that its conditions are weaker than those of existing frequentist approaches, in particular allowing a fixed number of changed components and requiring no sparsity of precision matrices.

Load-bearing premise

The theorems guarantee consistency only for hyperparameter $\alpha$ satisfying explicit inequalities, but the implemented method selects $\alpha$ by simulating datasets from the sample mean and covariance and picking the value whose empirical false positive rate is below 0.05; the paper does not prove that this data-driven $\alpha$ satisfies the theoretical conditions, so the practical procedure may run outside the regime where consistency and near-optimality are proven.

Editorial extensions

If this is right

For mean changes, consistent localization within $n_w$ is guaranteed provided each change point is separated by at least $n_w$ and some coordinate moves by at least $C \log(n_w \vee p)/n_w$ in standardized units; the number of changed coordinates may be fixed, which existing sparse-alternative tests do not allow.
The localization error $n_w/n$ is minimax up to a logarithmic factor for single mean changes and for covariance changes under the stated signal classes, so the chosen window size directly sets the time resolution one can promise.
Covariance detection requires no sparsity of the precision matrix and no irrepresentability condition, unlike earlier tests that estimated sparse precisions.
The multiscale majority-vote version stabilizes detection across window choices, provided at least half of the window sizes satisfy the theorem conditions and the window ladder grows geometrically.
On real data, running the covariance detector first and then the mean detector within segments yields change points that align with known financial and genomic events in the two reported applications.

Reading between the lines

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

The modularization into per-coordinate or per-pair Bayes factors suggests the same windowed maximum test could be applied to sub-Gaussian or heavy-tailed data after robustifying the variance estimators, but the paper's proofs rely on exact Gaussian likelihoods, so the calibration of the log Bayes factor outside normality is unverified.
Because the theory only requires one coordinate or pair with signal above the threshold, the method is intrinsically a maximum-type detector; the paper's own discussion notes it may miss many small coordinated changes, so a hybrid that first aggregates weak signals and then applies mxPBF is a natural testable extension.
The FPR-based $\alpha$ calibration simulates from the plug-in Gaussian fit; a reader who wants to deploy the method should treat the consistency theorems as conditional on $\alpha$ satisfying the stated inequalities and verify the calibration on their own null model.
The multiscale majority rule is proven only for window ladders with $n_w^{(r)} \le n_w^{(r+1)}/2$ and at least half of the windows meeting the theorem conditions; in practice, the stability of detected change points across window ladders could serve as a data-driven check.

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.

Desk Editor's Note

read the letter

Here's the short version: the paper delivers a real contribution—the first Bayesian change point detection method for high-dimensional mean and covariance that comes with consistency and near-minimax localization guarantees. The multiscale majority rule is a sensible practical fix for window choice, the simulations are extensive, and the code is on GitHub. If you work in this area, it is worth your time.

The main problem is a gap between theory and practice. Theorems 2.2 and 3.2 require a fixed alpha satisfying (7) or alpha > 6C3, but the implemented procedure selects alpha by FPR calibration on datasets simulated from the sample mean and covariance of the observed sequence, as described in Section 4.2 and Appendix B. No result shows that this data-dependent alpha satisfies the theorem conditions, and when change points exist the calibration null can be far from the true null because the pooled covariance is inflated by between-segment differences. So the consistency and localization guarantees apply to a version of the method that is not the one run in the simulations or applications. This is a load-bearing gap, not a cosmetic one.

What the paper does well beyond the novelty: the proof strategy is internally consistent, the lower-bound arguments are standard and clean, and the comparisons, while not always apples-to-apples, are honestly reported—the authors explicitly note where direct comparison is difficult. The delegation of tail bounds to Lee et al. (2024) is acceptable but means the paper is not self-contained; a reader has to trust that prior work. The claim of "much milder" conditions is somewhat overstated because the signal is measured in a different norm (elementwise maximum vs. l2), so the comparison is not directly against existing results.

Minor issues: the simulation figures have no error bars despite being based on 50 datasets, and the threshold Ccp is treated as a fixed default without sensitivity analysis. These are small.

This paper deserves a serious referee. The central idea is fresh, the method is implementable, and the gap is addressable: either prove the FPR-calibrated alpha satisfies the theorem conditions with high probability, or state the theorems for the data-dependent alpha directly. I would like to see this fixed before publication, but I would not desk reject it.

Referee Report

4 major / 6 minor

Summary. The paper proposes new Bayesian tests and change-point estimators for high-dimensional Gaussian sequences, using maximum pairwise Bayes factors computed in local windows. For mean changes it establishes consistency of the mxPBF test (Theorem 2.1), consistent estimation of the number and locations with error nw (Theorem 2.2), and a minimax lower bound showing near-optimal localization (Theorem 2.3). Analogous results are given for covariance changes (Theorems 3.1-3.4) under conditions (A2)-(A3*). The paper also proposes a multiscale majority-voting implementation, reports simulations comparing with Dette, Geomcp, Inspect, and Edivisive, and applies the methods to CGH and S&P 500 data. The theoretical results are stated for fixed alpha satisfying condition (7) or alpha > 6C3, while the implementation selects alpha by empirical false-positive calibration.

Significance. If the theorems are correct, the contributions are significant: they appear to be the first Bayesian change-point methods for high-dimensional mean and covariance structures with consistency under weaker sparsity and segment-length conditions than several frequentist competitors, and with near-minimax localization rates. The paper also includes explicit minimax lower-bound constructions, a public GitHub implementation, and comparisons with four existing methods. The modularized pairwise-Bayes-factor construction is computationally attractive because it avoids p-by-p matrix inversions. However, the main theorems are not self-contained, and the implemented tuning procedure is not covered by the theoretical guarantees; these issues must be addressed before the headline claims can be accepted.

major comments (4)

[Section 4.2 and Appendix B; Theorems 2.1-2.2 and 3.1-3.2] The implemented method is not the version covered by the main theorems. Theorems 2.1, 2.2, 3.1, and 3.2 require a fixed alpha satisfying condition (7) in the mean case or alpha > 6C3 in the covariance case, whereas Section 4.2 and Appendix B select alpha by empirical false-positive calibration: N=300 datasets are simulated from N_p(mu-hat, Sigma-hat), where mu-hat and Sigma-hat are the sample mean and covariance of the observed data, and alpha is chosen so that the simulated mxPBF exceeds C_th with frequency below 0.05. No result in the paper shows that this data-dependent alpha-hat satisfies condition (7) or alpha-hat > 6C3, nor that simulating from the plug-in N_p(mu-hat, Sigma-hat) is a valid null distribution when change points are present; in the covariance-change case the pooled Sigma-hat is not the null covariance. A parallel gap exists for the covariance prior constants a0, b0, b01, b02, which are fixed at 0.01 in Section 4.2 without checking the inequalities in conditions (A3) and (A3*). Consequently, the Abstract's consistency and near-optimal localization claims apply to a version of the algorithm that is not the one run in the simulations or applied to the real datasets.
[Appendix D, proofs of Theorems 2.1 and 3.1] The key probability bounds are not proved or stated in the paper. The proofs repeatedly invoke 'the proof of Theorem 2.1 in Lee et al. (2024)' or 'the proof of Theorem 3.1 in Lee et al. (2024)' to obtain tail bounds of the form (nw or p)^{-C} for the pairwise Bayes factors. These tail bounds are exactly what powers the null and alternative consistency arguments, and they are the source of the constants C1, C2, C3, Cbm, Cbm,a and of condition (7). As written, the main theorems are not self-contained: a reader cannot verify that the quoted lemmas hold under the paper's priors, windowing setup, and high-dimensional scaling without consulting an external manuscript. The needed lemmas should be stated and proved, or at least precisely stated with assumptions sufficient for the present paper.
[Proof of Theorem 3.2, Appendix D.2] The final bound in the proof of Theorem 3.2 says that a sum over K0 change points is at most 14 K0 (nw or p)^{-C} = o(1) for some constant 2 < C < C1. But Theorem 3.2 defines C1 as a constant arbitrarily close to, but slightly larger than, 1, so the interval (2, C1) is empty and the inequality is vacuous. The preceding step also needs an explicit relationship between C and C1,low to justify the o(1) conclusion from the assumption K0(nw or p)^{-C1,low} = o(1). This makes the proof of Theorem 3.2 invalid as written, although the theorem may be salvageable with corrected constant comparisons.
[Section 4.1, multiscale method] The multiscale 'mxPBF major' procedure is the method used in all simulations and real-data applications, but its theoretical justification is only a heuristic sentence: if at least the window sizes n_w^(1), ..., n_w^((|N|+1)/2) satisfy the conditions of Theorems 2.2 and 3.2 and n_w^(r) <= n_w^(r+1)/2, then 'it can be shown' that the localization rate is at most n_w^((|N|+1)/2)/n. No theorem, lemma, or proof is supplied for the majority-voting interval grouping, so the finite-sample claims and the reported F1/Hausdorff performance are not connected to the proved localization rates. The inconsistency between the theoretical procedure (single window size, fixed alpha) and the evaluated procedure (multiscale majority voting, calibrated alpha) should be resolved explicitly.

minor comments (6)

[Section 2.1, after equation (3)] The sentence 'If H_0 is true, we say that l is a change point' is backwards: under the equality null there is no change at l; the intended statement is presumably that if H_1 is true, l is a change point.
[Theorem 3.1(ii)] The condition '(i_k - i_{k-1}) or (i_{k+1} - i_k) >= n_w' uses a maximum where the analogous mean-structure condition in Theorem 2.1 uses a minimum; since the mxPBF at l = i_k uses n_w observations from each side, the minimum (both segment lengths at least n_w) appears to be what is needed.
[Section 4.5] There is a duplicated word in 'We define Delta = Delta1 + delta I_p, where where delta = |lambda_min(Delta1)| + 0.05'; the second 'where' should be removed.
[Section 5, real data analysis] The covariance detection method is applied after locally centering the data with a moving-average window, but Theorems 3.1-3.4 assume the data have mean exactly zero; no theory is given for the effect of this centering on the null distribution or on the localization guarantees.
[Figures 1-6 and Appendix A] Simulation results are reported as point summaries over only 50 datasets without error bars or confidence bands; given the visible variability in F1 scores and Hausdorff distances, reporting variability or replication counts would strengthen the empirical comparisons.
[Appendix B.1] The word 'controlls' in 'which controlls the empirical false positive rate at 0.05' is a typo; it should be 'controls'.

Circularity Check

0 steps flagged · score 0.0 of 10

No definitional circularity: consistency/localization proofs rely on independent two-sample PBF tail bounds from Lee et al. (2024) and in-text lower-bound constructions; the FPR-based alpha calibration is an unproven applicability gap, not a circular reduction.

full rationale

The load-bearing derivation is not circular. Theorem 2.2 (and 3.2) reduces consistent localization to the per-window mxPBF tail behavior established in Theorem 2.1 (and 3.1); those tail bounds are imported from the previously published two-sample testing analysis in Lee et al. (2024), which does not assume any change point result. The minimax lower bounds (Theorems 2.3, 3.3, 3.4) are proved in the paper by standard Le Cam constructions adapted from Wang and Samworth (2018) and Wang et al. (2021), with citations. The claimed near-optimality is a genuine rate match: when the signal is at the lower-bound threshold psi^2_min asymptotically log(nw or p)/nw, the nw/n upper localization rate is within a log factor of the 1/(n*psi^2_min) lower bound; this is minimax matching, not equivalence-by-definition. The one substantive weakness is a coverage gap, not circularity: Section 4.2 and Appendix B select alpha by FPR calibration using datasets simulated from N_p(mu_hat, Sigma_hat), whereas Theorems 2.1 and 3.1 require a fixed alpha satisfying (7) or alpha > 6*C3, and no result shows the calibrated alpha_hat meets those conditions or that simulation from the pooled sample moments preserves the null distribution when change points exist. Similarly, the multiscale localization claim in Section 4.1 is asserted with 'it can be shown' and no proof. Both are missing-support or robustness issues; neither makes a predicted quantity equal to an input by construction, because the consistency theorems apply to a well-defined fixed-alpha procedure and the calibration is an extra heuristic layered on top of it.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The central claim rests on a small set of tuning parameters, on Gaussian modeling assumptions, and on tail bounds imported from earlier papers by the same research group. No new latent variables, particles or physical entities are introduced; the method uses observed variables only.

free parameters (4)

alpha (prior concentration exponent) = selected by FPR calibration; grid 0.01 to 15
Controls prior variance via gamma=(nw or p)^(-alpha); theory requires condition (7) or alpha > 6*C3, while practice sets it by empirical false positive rate with no proof that the selected value satisfies the theorem conditions.
window size nw (or window set N) = N={25,60,100} in simulations
Consistency and localization require nw <= delta_n and condition (A1); the optimal nw depends on unknown signal strength and change point spacing. Multiscale aggregation reduces but does not eliminate this dependency.
threshold Ccp = 10 by default
Test and estimation decisions compare the max PBF to Ccp; theory requires a fixed threshold but gives no guidance, and finite-sample behavior depends on this arbitrary choice.
covariance prior shape and rate a0, b0, b01, b02 = 0.01
Chosen by hand to minimize their influence; these hyperparameters enter the covariance PBF and appear in condition (A3*).

assumptions (5)

domain assumption Tail probability bounds for the pairwise Bayes factors from Lee et al. (2024) hold as stated.
Theorems 2.1 and 3.1 in this paper are proved by direct appeal to the proofs of Theorem 2.1 and 3.1 of Lee et al. (2024); the present paper does not derive these bounds.
domain assumption Data are independent Gaussian draws: N_p(mu_i, Sigma) for the mean case and N_p(0, Sigma_i) for the covariance case.
The likelihood, PBF formulas and all theorems in Sections 2 and 3 assume Gaussianity, with a common covariance in the mean case and zero mean in the covariance case.
domain assumption Window size satisfies (A1) with nw <= n/2 and log(nw or p)/nw = o(1), and for estimation nw <= delta_n.
Required for the null bound and for separation of change points; used in Theorems 2.1 through 2.3 and 3.1 through 3.4.
ad hoc to paper The FPR calibration simulates a valid null distribution from N_p(muhat, Sigmahat).
Appendix B generates null datasets from the sample mean and covariance of the observed data to pick alpha; no theorem shows this preserves the nominal false positive rate or satisfies the theoretical condition on alpha.
domain assumption The true number of change points K0 is fixed, and the minimum spacing delta_n and signal strength conditions hold.
Estimation theorems require K0(nw or p)^(-C1low)=o(1), delta_n >= nw and lower bounds on psi_min or covariance signal; these are not verified from data.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Bayesian optimal change point detection in high-dimensions." pith.science (2026). https://pith.science/paper/73H5PI7K

@misc{pith2026241114864,
  author       = {Pith},
  title        = {Pith review of: Bayesian optimal change point detection in high-dimensions},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/73H5PI7K}},
  note         = {Machine review of arXiv:2411.14864}
}

read the original abstract

We propose the first Bayesian methods for detecting change points in high-dimensional mean and covariance structures. These methods are constructed using pairwise Bayes factors, leveraging modularization to identify significant changes in individual components efficiently. We establish that the proposed methods consistently detect and estimate change points under much milder conditions than existing approaches in the literature. Additionally, we demonstrate that their localization rates are nearly optimal in terms of rates. The practical performance of the proposed methods is evaluated through extensive simulation studies, where they are compared to state-of-the-art techniques. The results show comparable or superior performance across most scenarios. Notably, the methods effectively detect change points whenever signals of sufficient magnitude are present, irrespective of the number of signals. Finally, we apply the proposed methods to genetic and financial datasets, illustrating their practical utility in real-world applications.

Figures

Figures reproduced from arXiv: 2411.14864 by the authors.

**Figure 2.** F1 scores and Hausdorff distances for change point detection in mean structure based on [PITH_FULL_IMAGE:figures/full_fig_p021_2.png] view at source ↗

**Figure 3.** F1 scores and Hausdorff distances for change point detection in mean structures based [PITH_FULL_IMAGE:figures/full_fig_p022_3.png] view at source ↗

Figures from the paper (22 more)

**Figure 4.** Figure 4: F1 scores for change point detection in covariance structure based on 50 simulated datasets [PITH_FULL_IMAGE:figures/full_fig_p024_4.png]

**Figure 5.** Figure 5: F1 scores and Hausdorff distances for change point detection in covariance structure based [PITH_FULL_IMAGE:figures/full_fig_p025_5.png]

**Figure 6.** Figure 6: F1 scores and Hausdorff distances for change point detection in covariance structure based [PITH_FULL_IMAGE:figures/full_fig_p026_6.png]

**Figure 7.** Figure 7: Log-intensity ratio measurements from microarray data of 43 individuals, with vertical [PITH_FULL_IMAGE:figures/full_fig_p028_7.png]

**Figure 8.** Figure 8: Log returns of 479 companies within the S&P 500 are displayed, with vertical dotted lines [PITH_FULL_IMAGE:figures/full_fig_p029_8.png]

**Figure 9.** Figure 9: F1 scores for change point detection in mean structure based on 50 simulated datasets [PITH_FULL_IMAGE:figures/full_fig_p031_9.png]

**Figure 10.** Figure 10: F1 scores and Hausdorff distances for change point detection in mean structure based [PITH_FULL_IMAGE:figures/full_fig_p032_10.png]

**Figure 11.** Figure 11: F1 scores and Hausdorff distances for change point detection in mean structure based [PITH_FULL_IMAGE:figures/full_fig_p033_11.png]

**Figure 12.** Figure 12: F1 scores and Hausdorff distances based for change point detection in mean structure [PITH_FULL_IMAGE:figures/full_fig_p035_12.png]

**Figure 13.** Figure 13: F1 scores and Hausdorff distances for change point detection in mean structure based [PITH_FULL_IMAGE:figures/full_fig_p036_13.png]

**Figure 14.** Figure 14: F1 scores for change point detection in covariance structure based on 50 simulated [PITH_FULL_IMAGE:figures/full_fig_p037_14.png]

**Figure 15.** Figure 15: F1 scores and Hausdorff distances for change point detection in covariance structure [PITH_FULL_IMAGE:figures/full_fig_p038_15.png]

**Figure 16.** Figure 16: F1 scores and Hausdorff distances for change point detection in covariance structure [PITH_FULL_IMAGE:figures/full_fig_p039_16.png]

**Figure 17.** Figure 17: F1 scores and Hausdorff distances for change point detection in covariance structure [PITH_FULL_IMAGE:figures/full_fig_p040_17.png]

**Figure 18.** Figure 18: F1 scores and Hausdorff distances for change point detection in covariance structure [PITH_FULL_IMAGE:figures/full_fig_p041_18.png]

**Figure 19.** Figure 19: The selected ˆα, which controlls the empirical false positive rate at 0.05, is based on 100 replications for both the mean and covariance methods, and for each of the following hypotheses: H0 and H1,M, with p = 100. C The multiscale method In this section, we compare …

**Figure 20.** Figure 20: F1 scores for change point detection in mean structure are based on 50 simulated [PITH_FULL_IMAGE:figures/full_fig_p044_20.png]

**Figure 21.** Figure 21: F1 scores for change point detection in mean structure based on 50 simulated datasets [PITH_FULL_IMAGE:figures/full_fig_p045_21.png]

**Figure 22.** Figure 22: F1 scores for change point detection in mean structure based on 50 simulated datasets [PITH_FULL_IMAGE:figures/full_fig_p046_22.png]

**Figure 23.** Figure 23: F1 scores for change point detection in covariance structure based on 50 simulated [PITH_FULL_IMAGE:figures/full_fig_p046_23.png]

**Figure 24.** Figure 24: F1 scores for change point detection in covariance structure based on 50 simulated [PITH_FULL_IMAGE:figures/full_fig_p047_24.png]

**Figure 25.** Figure 25: F1 scores for change point detection in covariance structure based on 50 simulated [PITH_FULL_IMAGE:figures/full_fig_p048_25.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

5 extracted references · 3 canonical work pages

[1]

and Buzun, N

Avanesov, V. and Buzun, N. (2018). Change-point detection in high-dimensional covariance struc- ture, Electronic Journal of Statistics 12(2): 3254–3294. Berthet, Q. and Rigollet, P. (2013). Optimal detection of sparse principal components in high dimension, The Annals of Statistics 41(4): 1780–1815. Bleakley, K. and Vert, J.-P. (2011). The group fused las...

arXiv 2018
[83]

Yu, B. (1997). Assouad, fano, and le cam, Festschrift for Lucien Le Cam , Springer, pp. 423–435. Zhang, N. R. and Siegmund, D. O. (2007). A modified bayes information criterion with applications to the analysis of comparative genomic hybridization data, Biometrics 63(1): 22–32. 56

work page 1997
[454]

and Harchaoui, Z

Enikeeva, F. and Harchaoui, Z. (2019). High-dimensional change-point detection under sparse alternatives, The Annals of Statistics 47(4): 2051–2079. Foret, P. (2019). Sp500r: Easy loading of sp500 stocks data. Github R package version 0.1.0. Fryzlewicz, P. (2014). Wild binary segmentation for multiple change-point detection, The Annals of Statistics 42(6)...

work page 2019
[507]

disorder

Van den Burg, G. J. and Williams, C. K. (2020). An evaluation of change point detection algorithms, arXiv preprint arXiv:2003.06222 . Vostrikova, L. J. (1981). Detecting “disorder” in multidimensional random processes., Soviet Math- ematics Doklady 24: 55–59. Wang, D., Yu, Y. and Rinaldo, A. (2021). Optimal covariance change point localization in high dim...

arXiv 2020
[4419]

and Lin, L

Lee, K., You, K. and Lin, L. (2024). Bayesian optimal two-sample tests for high-dimensional gaussian populations, Bayesian Analysis 19(3): 869–893. Matteson, D. S. and James, N. A. (2014). A nonparametric approach for multiple change point analysis of multivariate data, Journal of the American Statistical Association 109(505): 334–345. 55 Pollard, D. (200...

work page 2024

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.

[1] [1]

and Buzun, N

Avanesov, V. and Buzun, N. (2018). Change-point detection in high-dimensional covariance struc- ture, Electronic Journal of Statistics 12(2): 3254–3294. Berthet, Q. and Rigollet, P. (2013). Optimal detection of sparse principal components in high dimension, The Annals of Statistics 41(4): 1780–1815. Bleakley, K. and Vert, J.-P. (2011). The group fused las...

arXiv 2018

[2] [83]

Yu, B. (1997). Assouad, fano, and le cam, Festschrift for Lucien Le Cam , Springer, pp. 423–435. Zhang, N. R. and Siegmund, D. O. (2007). A modified bayes information criterion with applications to the analysis of comparative genomic hybridization data, Biometrics 63(1): 22–32. 56

work page 1997

[3] [454]

and Harchaoui, Z

Enikeeva, F. and Harchaoui, Z. (2019). High-dimensional change-point detection under sparse alternatives, The Annals of Statistics 47(4): 2051–2079. Foret, P. (2019). Sp500r: Easy loading of sp500 stocks data. Github R package version 0.1.0. Fryzlewicz, P. (2014). Wild binary segmentation for multiple change-point detection, The Annals of Statistics 42(6)...

work page 2019

[4] [507]

disorder

Van den Burg, G. J. and Williams, C. K. (2020). An evaluation of change point detection algorithms, arXiv preprint arXiv:2003.06222 . Vostrikova, L. J. (1981). Detecting “disorder” in multidimensional random processes., Soviet Math- ematics Doklady 24: 55–59. Wang, D., Yu, Y. and Rinaldo, A. (2021). Optimal covariance change point localization in high dim...

arXiv 2020

[5] [4419]

and Lin, L

Lee, K., You, K. and Lin, L. (2024). Bayesian optimal two-sample tests for high-dimensional gaussian populations, Bayesian Analysis 19(3): 869–893. Matteson, D. S. and James, N. A. (2014). A nonparametric approach for multiple change point analysis of multivariate data, Journal of the American Statistical Association 109(505): 334–345. 55 Pollard, D. (200...

work page 2024