pith. machine review for the scientific record. sign in

arxiv: 2604.14182 · v1 · submitted 2026-03-31 · 📊 stat.ME · stat.ML

Recognition: 2 theorem links

· Lean Theorem

Cellwise Outliers

Jakob Raymaekers, Mia Hubert, Peter J. Rousseeuw

Authors on Pith no claims yet

Pith reviewed 2026-05-13 23:42 UTC · model grok-4.3

classification 📊 stat.ME stat.ML
keywords cellwise outliersrobust statisticshigh-dimensional dataoutlier detectioncovariance estimationregressionprincipal component analysismissing values
0
0 comments X

The pith

Cellwise outliers as individual bad entries can contaminate over half the cases in high-dimensional data, requiring robust methods that differ from traditional casewise approaches.

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

The paper shows that the usual definition of an outlier as a whole case no longer suffices once data matrices grow high-dimensional. A modest number of anomalous individual cells can now spread across and spoil a majority of rows, so methods that treat entire cases as the unit of analysis lose effectiveness. Detecting these cellwise outliers and building estimators that remain reliable around them calls for techniques that drop some familiar equivariance properties and instead operate directly on the cells. The review surveys ten years of progress on location and covariance estimation, regression, principal component analysis, tensor data, and related problems, noting that the resulting cellwise procedures also tend to accommodate missing values without extra machinery.

Core claim

Traditional outlier detection treats an entire case as the basic unit, but in high-dimensional data even a small fraction of anomalous cells can contaminate more than half the observations. Robust methods must therefore target individual cells rather than whole cases, which means relinquishing some intuitive equivariance properties that casewise methods rely on. Over the past decade this shift has produced workable procedures for estimating location and covariance, for regression, for principal component analysis, and for tensor data; these cellwise techniques are becoming the standard choice for high-dimensional problems and routinely handle missing values at the same time.

What carries the argument

Cellwise outliers, defined as anomalous single entries inside a data matrix or tensor, which force the construction of robust estimators that act on cells instead of entire cases and therefore relax certain equivariance requirements.

Load-bearing premise

A relatively small proportion of outlying cells can contaminate over half the cases, making casewise methods insufficient for modern high-dimensional data.

What would settle it

A controlled experiment on high-dimensional data with a known small percentage of contaminated cells that nonetheless affect most rows, in which standard casewise robust estimators recover the true parameters as accurately as cellwise estimators, would undermine the claim that cellwise methods are required.

Figures

Figures reproduced from arXiv: 2604.14182 by Jakob Raymaekers, Mia Hubert, Peter J. Rousseeuw.

Figure 1
Figure 1. Figure 1: Illustration of the casewise, cellwise, and mixed contamination models. Black means [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Left: some variables of the Top Gear data, illustrating that cellwise outliers need not [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Illustration of out-of-sample prediction with the cellLTS method. [PITH_FULL_IMAGE:figures/full_fig_p016_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Upper panel: Illustration of cellPCA imputation for [PITH_FULL_IMAGE:figures/full_fig_p022_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Some frames of the Dog Walker data, and the corresponding ROMPCA cellmaps. The [PITH_FULL_IMAGE:figures/full_fig_p026_5.png] view at source ↗
read the original abstract

In statistics and machine learning, the traditional meaning of the terms `outlier' and `anomaly' is a case in the dataset that behaves differently from the bulk of the data. This raises suspicion that it may belong to a different population. But nowadays increasing attention is being paid to so-called cellwise outliers. These are individual values somewhere in the data matrix (or data tensor). Depending on the dimension, even a relatively small proportion of outlying cells can contaminate over half the cases, which is a problem for existing casewise methods. It turns out that detecting cellwise outliers as well as constructing cellwise robust methods requires techniques that are quite different from the casewise setting. For instance, one has to let go of some intuitive equivariance properties. The problem is difficult, but the past decade has seen substantial progress. For high-dimensional data the cellwise approach is becoming dominant, and typically can deal with missing values as well. We review developments in the estimation of location and covariance matrices as well as regression methods, principal component analysis, methods for tensor data, and various other settings.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript is a review of developments in cellwise outlier detection and robust methods for data matrices and tensors. It contrasts cellwise outliers (individual aberrant entries) with traditional casewise outliers, notes that under an independent contamination model even modest cellwise contamination fractions can affect over half the observations in high dimensions, and argues that cellwise techniques require different tools (including relaxation of some equivariance properties). The review covers progress in location and covariance estimation, regression, PCA, tensor methods, and related settings, concluding that cellwise approaches are becoming dominant for high-dimensional data and often accommodate missing values.

Significance. As a timely synthesis of the literature, the review usefully documents the shift toward cellwise robust procedures in high-dimensional statistics and their compatibility with missing-data handling. If the cited developments are accurately summarized, the paper provides a consolidated reference that can orient researchers to the key distinctions, technical challenges, and available methods in this area.

major comments (2)
  1. [Abstract / Introduction] The central contamination claim (small cellwise fraction contaminating >50% of cases) is stated in the abstract and introduction but would benefit from an explicit short derivation or citation to the independent contamination model (1-α)^p in the opening section, to make the quantitative motivation self-contained for readers unfamiliar with the model.
  2. [Conclusion / Summary of methods] The statement that cellwise methods 'typically can deal with missing values as well' is asserted without a dedicated subsection or table summarizing which reviewed methods (location, regression, PCA, tensor) explicitly handle missingness and under what assumptions; this weakens the comparative claim.
minor comments (2)
  1. [Abstract] The abstract introduces the term 'anomaly' alongside 'outlier' but does not clarify whether the cellwise framework treats them identically or distinguishes them; a brief sentence would improve precision.
  2. [Introduction] Notation for cellwise contamination probability (α) and dimension (p) is used informally in the text; consistent definition in a preliminary section would aid readability.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the positive evaluation of our manuscript and for the helpful suggestions. We address each major comment below.

read point-by-point responses

  1. Referee: [Abstract / Introduction] The central contamination claim (small cellwise fraction contaminating >50% of cases) is stated in the abstract and introduction but would benefit from an explicit short derivation or citation to the independent contamination model (1-α)^p in the opening section, to make the quantitative motivation self-contained for readers unfamiliar with the model.

    Authors: We agree with this suggestion. To make the motivation more self-contained, we will include a brief derivation of the independent contamination model in the introduction. Specifically, we will explain that under the model where each cell is contaminated independently with probability α, the probability that a given case remains uncontaminated is (1-α)^p, so the expected proportion of contaminated cases is 1-(1-α)^p. For example, with p=100 and α=0.01 this exceeds 0.63. We will also add a citation to the original reference for this model. revision: yes

  2. Referee: [Conclusion / Summary of methods] The statement that cellwise methods 'typically can deal with missing values as well' is asserted without a dedicated subsection or table summarizing which reviewed methods (location, regression, PCA, tensor) explicitly handle missingness and under what assumptions; this weakens the comparative claim.

    Authors: We appreciate this observation. In the revised manuscript, we will add a summary table (or a dedicated paragraph in the conclusion) that lists the main methods reviewed in each section (location/covariance, regression, PCA, tensors) and indicates whether they handle missing values, along with the underlying assumptions (e.g., missing at random). This will provide concrete support for the claim that cellwise approaches often accommodate missing data. revision: yes

Circularity Check

0 steps flagged

No significant circularity in this review paper

full rationale

This is a review summarizing external literature on cellwise outlier methods without presenting new derivations, equations, fitted parameters, or predictions. The central descriptive claim that cellwise approaches are becoming dominant for high-dimensional data follows from documented field progress and external citations rather than internal reduction. The contamination statement (small cellwise fraction contaminating over half the cases) is a direct mathematical consequence of the independent contamination model (1-α)^p dropping below 0.5 for large p, which requires no self-definition, self-citation chain, or ansatz from the present paper. No load-bearing steps reduce to the paper's own inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

As a review paper the work introduces no new free parameters, axioms, or invented entities; it summarizes techniques from the existing robust statistics literature.

pith-pipeline@v0.9.0 · 5488 in / 1029 out tokens · 55561 ms · 2026-05-13T23:42:30.705350+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

Reference graph

Works this paper leans on

90 extracted references · 90 canonical work pages · 1 internal anchor

  1. [1]

    Aerts, S. and I. Wilms (2017). Cellwise robust regularized discriminant analysis. Statistical Analysis and Data Mining: The ASA Data Science Journal\/ 10\/ (6), 436--447

    work page 2017

  2. [2]

    Leung, V

    Agostinelli, C., A. Leung, V. J. Yohai, and R. H. Zamar (2015). Robust estimation of multivariate location and scatter in the presence of cellwise and casewise contamination. TEST\/ 24 , 441--461

    work page 2015

  3. [3]

    Alfons, A. (2022). robustHD : Robust methods for high-dimensional data. R package, https://CRAN.R-project.org/package=robustHD, DOI 10.32614/CRAN.package.robustHD

    work page doi:10.32614/cran.package.robusthd 2022

  4. [4]

    Van Aelst, V

    Alqallaf, F., S. Van Aelst, V. J. Yohai, and R. H. Zamar (2009). Propagation of outliers in multivariate data. The Annals of Statistics\/ 37 , 311--331

    work page 2009

  5. [5]

    Ballard, G. and T. G. Kolda (2025). Tensor Decompositions for Data Science . Cambridge University Press

    work page 2025

  6. [6]

    Hathaway, R

    Bezdek, J., R. Hathaway, R. Howard, C. Wilson, and M. Windham (1987). Local convergence analysis of a grouped variable version of coordinate descent. Journal of Optimization Theory and Applications\/ 54 , 471--477

    work page 1987

  7. [7]

    Bi, X., X. Tang, Y. Yuan, Y. Zhang, and A. Qu (2021). Tensors in statistics. Annual Review of Statistics and Its Application\/ 8 , 345--368

    work page 2021

  8. [8]

    Croux, and I

    Bottmer, L., C. Croux, and I. Wilms (2022). Sparse regression for large data sets with outliers. European Journal of Operational Research\/ 297\/ (2), 782--794

    work page 2022

  9. [9]

    Cornelissen, and C

    Boudt, K., J. Cornelissen, and C. Croux (2012). The gaussian rank correlation estimator: robustness properties. Statistics and Computing\/ 22\/ (2), 471--483

    work page 2012

  10. [10]

    Hubert, B

    Centofanti, F., M. Hubert, B. Palumbo, and P. J. Rousseeuw (2025). Multivariate singular spectrum analysis by robust diagonalwise low-rank approximation. Journal of Computational and Graphical Statistics\/ 34 , 360--373

    work page 2025

  11. [11]

    Hubert, and P

    Centofanti, F., M. Hubert, and P. J. Rousseeuw (2025). Cellwise and casewise robust covariance in high dimensions. arXiv preprint, https://arxiv.org/abs/2505.19925 \/

    work page arXiv 2025

  12. [12]

    Hubert, and P

    Centofanti, F., M. Hubert, and P. J. Rousseeuw (2026). Robust principal components by casewise and cellwise weighting. Technometrics, to appear \/ \!\! , \; https://doi.org/10.1080/00401706.2026.2643216\,

    work page doi:10.1080/00401706.2026.2643216 2026

  13. [13]

    Pyneeandee, and G

    Christidis, A., J. Pyneeandee, and G. Cohen-Freue (2026). Fast and Scalable Cellwise-Robust Ensembles for High-Dimensional Data . arXiv preprint, https://arxiv.org/abs/2603.20940 \/

    work page arXiv 2026

  14. [14]

    Filzmoser, and H

    Croux, C., P. Filzmoser, and H. Fritz (2013). Robust S parse P rincipal C omponent A nalysis. Technometrics\/ 55 , 202--214

    work page 2013

  15. [15]

    Filzmoser, and M

    Croux, C., P. Filzmoser, and M. Oliveira (2007). Algorithms for P rojection- P ursuit R obust P rincipal C omponent A nalysis. Chemometrics and Intelligent Laboratory Systems\/ 87 , 218--225

    work page 2007

  16. [16]

    Filzmoser, G

    Croux, C., P. Filzmoser, G. Pison, and P. J. Rousseeuw (2003). Fitting multiplicative models by robust alternating regressions. Statistics and Computing\/ 13 , 23--36

    work page 2003

  17. [17]

    Croux, C. and V. \"O llerer (2016). Robust and sparse estimation of the inverse covariance matrix using rank correlation measures. In Recent Advances in Robust Statistics: Theory and Applications , pp.\ 35--55. Springer

    work page 2016

  18. [18]

    Danilov, M. (2010). Robust estimation of multivariate scatter in non-affine equivariant scenarios . Ph.\ D. thesis, Statistics Dept

    work page 2010

  19. [19]

    Danilov, M., V. J. Yohai, and R. H. Zamar (2012). Robust estimation of multivariate location and scatter in the presence of missing data. Journal of the American Statistical Association\/ 107 , 1178--1186

    work page 2012

  20. [20]

    De La Torre, F. and M. J. Black (2003). A framework for robust subspace learning. International Journal of Computer Vision\/ 54 , 117--142

    work page 2003

  21. [21]

    Donoho, D. and P. Huber (1983). The notion of breakdown point. In P. Bickel, K. Doksum, and J. Hodges (Eds.), A Festschrift for Erich Lehmann , Belmont, pp.\ 157--184. Wadsworth

    work page 1983

  22. [22]

    Donoho, D. L. (1982). Breakdown properties of multivariate location estimators. P h D qualifying paper, H arvard U niversity

    work page 1982

  23. [23]

    Engelen, S. and M. Hubert (2011). Detecting outlying samples in a parallel factor analysis model. Analytica Chimica Acta\/ 705 , 155--165

    work page 2011

  24. [24]

    Midtgaard, and S

    Esbensen, K., T. Midtgaard, and S. Sch\"onkopf (1996). Multivariate Analysis in Practice: A Training Package . Camo As, Oslo

    work page 1996

  25. [25]

    Farcomeni, A. (2014). Robust constrained clustering in presence of entry-wise outliers. Technometrics\/ 56\/ (1), 102--111

    work page 2014

  26. [26]

    Farcomeni, A. and A. Leung (2019). Package snipEM: Snipping Methods for Robust Estimation and Clustering . CRAN, R package version 1.0.1

    work page 2019

  27. [27]

    Hastie, and R

    Friedman, J., T. Hastie, and R. Tibshirani (2008). Sparse inverse covariance estimation with the graphical lasso. Biostatistics\/ 9\/ (3), 432--441

    work page 2008

  28. [28]

    Rivera-Garc \' a, A

    Garc \' a-Escudero, L.-A., D. Rivera-Garc \' a, A. Mayo-Iscar, and J. Ortega (2021). Cluster analysis with cellwise trimming and applications for the robust clustering of curves. Information Sciences\/ 573 , 100--124

    work page 2021

  29. [29]

    Gervini, D. and V. J. Yohai (2002). A class of robust and fully efficient regression estimators. The Annals of Statistics\/ 30\/ (2), 583--616

    work page 2002

  30. [30]

    Gnanadesikan, R. and J. R. Kettenring (1972). Robust estimates, residuals, and outlier detection with multiresponse data. Biometrics\/ 28 , 81--124

    work page 1972

  31. [31]

    Grygar, T. M., U. Radojičić, I. Pavlů, S. Greven, J. G. Nešlehová, Štěpánka Tůmová, and K. Hron (2024). Exploratory functional data analysis of multivariate densities for the identification of agricultural soil contamination by risk elements. Journal of Geochemical Exploration\/ 259 , 107416

    work page 2024

  32. [32]

    Hampel, F. R., E. M. Ronchetti, P. Rousseeuw, and W. A. Stahel (1986). Robust Statistics: the Approach based on Influence Functions . Wiley-Interscience, New York

    work page 1986

  33. [33]

    Hampel, F. R., P. J. Rousseeuw, and E. Ronchetti (1981). The Change-of-Variance Curve and Optimal Redescending M-Estimators . Journal of the American Statistical Association\/ 76\/ (375), 643--648

    work page 1981

  34. [34]

    Centofanti, M

    Hirari, M., F. Centofanti, M. Hubert, and S. Van Aelst (2026a). Casewise and cellwise robust multilinear principal component analysis. Journal of Computational and Graphical Statistics, to appear \/ \!\! , \;https://doi.org/10.1080/10618600.2026.2637632

    work page doi:10.1080/10618600.2026.2637632 2026

  35. [35]

    Centofanti, M

    Hirari, M., F. Centofanti, M. Hubert, and S. Van Aelst (2026b). Robust tensor-on-tensor regression. arXiv preprint, https://arxiv.org/abs/2603.25911 \/

    work page arXiv

  36. [36]

    Hubert, and P

    Hirari, M., M. Hubert, and P. J. Rousseeuw (2025). Graphical tools for visualizing cellwise and casewise outliers. Journal of Data Science, Statistics, and Visualisation,\/ 5\/ (10), DOI (https://doi.org/jdssv.v5i10.165)

    work page 2025

  37. [37]

    Huber, P. J. (1964). Robust Estimation of a Location Parameter . The Annals of Mathematical Statistics\/ 35\/ (1), 73 -- 101

    work page 1964

  38. [38]

    Hubert, M. and M. Hirari (2024). MacroPARAFAC for handling rowwise and cellwise outliers in incomplete multiway data. Chemometrics and Intelligent Laboratory Systems\/ 251 , 105170

    work page 2024

  39. [39]

    Reynkens, E

    Hubert, M., T. Reynkens, E. Schmitt, and T. Verdonck (2016). Sparse PCA for H igh- D imensional D ata W ith O utliers. Technometrics\/ 58 , 424--434

    work page 2016

  40. [40]

    Hubert, M., P. J. Rousseeuw, and P. Segaert (2015). Multivariate functional outlier detection (with discussion). Statistical Methods and Applications\/ 24 , 177--202

    work page 2015

  41. [41]

    Hubert, M., P. J. Rousseeuw, and W. Van den Bossche (2019). Macro PCA : An all-in-one PCA method allowing for missing values as well as cellwise and rowwise outliers. Technometrics\/ 61\/ (4), 459--473

    work page 2019

  42. [42]

    Hubert, M., P. J. Rousseeuw, and K. Vanden Branden (2005). ROBPCA : A new approach to robust principal component analysis. Technometrics\/ 47 , 64--79

    work page 2005

  43. [43]

    Van Kerckhoven, and T

    Hubert, M., J. Van Kerckhoven, and T. Verdonck (2012). Robust PARAFAC for incomplete data. Journal of Chemometrics\/ 26 , 290--298

    work page 2012

  44. [44]

    Hara, and K

    Inoue, K., K. Hara, and K. Urahama (2009). Robust multilinear principal component analysis. In 2009 IEEE 12th International Conference on Computer Vision , pp.\ 591--597

    work page 2009

  45. [45]

    Fujisawa, and M

    Katayama, S., H. Fujisawa, and M. Drton (2018). Robust and sparse G aussian graphical modelling under cell-wise contamination. Stat\/ 7\/ (1), e181

    work page 2018

  46. [46]

    Lee, H. Y., M. R. Gahrooei, H. Liu, and M. Pacella (2024). Robust tensor-on-tensor regression for multidimensional data modeling. IISE Transactions\/ 56\/ (1), 43--53

    work page 2024

  47. [47]

    Leggett, S. (2021). Migration and cultural integration in the early medieval cemetery of finglesham, kent, through stable isotopes. Archaeological and Anthropological Sciences\/ 13\/ (10), 171

    work page 2021

  48. [48]

    Danilov, V

    Leung, A., M. Danilov, V. Yohai, and R. Zamar (2019). GSE : Robust Estimation in the Presence of Cellwise and Casewise Contamination and Missing Data . CRAN, R package version 4.2, https://CRAN.R-project.org/package=GSE

    work page 2019

  49. [49]

    Yohai, and R

    Leung, A., V. Yohai, and R. Zamar (2017). Multivariate location and scatter matrix estimation under cellwise and casewise contamination. Computational Statistics & Data Analysis\/ 111 , 59--76

    work page 2017

  50. [50]

    Zhang, and R

    Leung, A., H. Zhang, and R. Zamar (2016). Robust regression estimation and inference in the presence of cellwise and casewise contamination. Computational Statistics & Data Analysis\/ 99 , 1--11

    work page 2016

  51. [51]

    Mukhopadhyay, and M

    Liu, A., R. Mukhopadhyay, and M. Markatou (2025). MDDC: An R and Python package for adverse event identification in pharmacovigilance data . Springer Nature Scientific Reports\/ 15 , 21317

    work page 2025

  52. [52]

    Marron, D

    Locantore, N., J. Marron, D. Simpson, N. Tripoli, J. Zhang, and K. Cohen (1999). Robust principal component analysis for functional data. Test\/ 8 , 1--73

    work page 1999

  53. [53]

    Lock, E. F. (2018). Tensor-on-tensor regression. Journal of Computational and Graphical Statistics\/ 27\/ (3), 638--647

    work page 2018

  54. [54]

    Loh, P.-L. and X. L. Tan (2018). High-dimensional robust precision matrix estimation: Cellwise corruption under -contamination. Electronic Journal of Statistics\/ 12 , 1429--1467

    work page 2018

  55. [55]

    Lopuha\"a, H. P. and P. J. Rousseeuw (1991). Breakdown points of affine equivariant estimators of multivariate location and covariance matrices. The Annals of Statistics\/ 19 , 229--248

    work page 1991

  56. [56]

    Lu, H., K. N. Plataniotis, and A. N. Venetsanopoulos (2008). MPCA : Multilinear principal component analysis of tensor objects. IEEE Transactions on Neural Networks\/ 19\/ (1), 18--39

    work page 2008

  57. [57]

    Maronna, R. (2005). Principal components and orthogonal regression based on robust scales. Technometrics\/ 47 , 264--273

    work page 2005

  58. [58]

    Maronna, R. A. and V. J. Yohai (2008). Robust low-rank approximation of data matrices with elementwise contamination. Technometrics\/ 50\/ (3), 295--304

    work page 2008

  59. [59]

    Radojičić, and P

    Mayrhofer, M., U. Radojičić, and P. Filzmoser (2025). Robust covariance estimation and explainable outlier detection for matrix-valued data. Technometrics\/ 67\/ (3), 516--530

    work page 2025

  60. [60]

    Alfons, and C

    \"O llerer, V., A. Alfons, and C. Croux (2016). The shooting S -estimator for robust regression. Computational Statistics\/ 31\/ (3), 829--844

    work page 2016

  61. [61]

    \"O llerer, V. and C. Croux (2015). Robust high-dimensional precision matrix estimation. In Modern Nonparametric, Robust and Multivariate methods , pp.\ 325--350. Springer

    work page 2015

  62. [62]

    Vana-Gür, and P

    Pfeiffer, P., L. Vana-Gür, and P. Filzmoser (2025). Cellwise robust and sparse principal component analysis. Advances in Data Analysis and Classification, to appear \/

    work page 2025

  63. [63]

    Walczak, and D

    Pravdova, V., B. Walczak, and D. Massart (2001). A robust version of the T ucker3 model. Chemometrics and Intelligent Laboratory Systems\/ 59 , 75--88

    work page 2001

  64. [64]

    Raymaekers, J. and P. J. Rousseeuw (2021a). Fast robust correlation for high-dimensional data. Technometrics\/ 63 , 184--198, https://doi.org/10.1080/00401706.2019.1677270

    work page doi:10.1080/00401706.2019.1677270 2019

  65. [65]

    Raymaekers, J. and P. J. Rousseeuw (2021b). Handling cellwise outliers by sparse regression and robust covariance. Journal of Data Science, Statistics, and Visualisation\/ 1\/ (3), DOI https://doi.org/10.52933/jdssv.v1i3.18)

    work page doi:10.52933/jdssv.v1i3.18

  66. [66]

    Raymaekers, J. and P. J. Rousseeuw (2022). cellWise : Analyzing data with cellwise outliers. R package, https://CRAN.R-project.org/package=cellWise, DOI 10.32614/CRAN.package.cellWise

    work page doi:10.32614/cran.package.cellwise 2022

  67. [67]

    Raymaekers, J. and P. J. Rousseeuw (2024a). The C ellwise M inimum C ovariance D eterminant estimator. Journal of the American Statistical Association\/ 119 , 2610--2621, DOI https://doi.org/10.1080/01621459.2023.2267777

    work page doi:10.1080/01621459.2023.2267777 2023

  68. [68]

    Raymaekers, J. and P. J. Rousseeuw (2024b). Transforming variables to central normality. Machine Learning\/ 113 , 4953–4975, DOI https://doi.org/10.1007/s10994--021--05960--5

    work page doi:10.1007/s10994--021--05960--5

  69. [69]

    Raymaekers, J. and P. J. Rousseeuw (2026a). Challenges of cellwise outliers. Econometrics and Statistics\/ 38 , 6--25, DOI https://doi.org/10.1016/j.ecosta.2024.02.002

    work page doi:10.1016/j.ecosta.2024.02.002 2024

  70. [70]

    Raymaekers, J. and P. J. Rousseeuw (2026b). Least trimmed squares regression with missing values and cellwise outliers. arXiv preprint, https://arxiv.org/abs/2603.04632 \/

    work page arXiv

  71. [71]

    Rousseeuw, P. J. (1984). Least median of squares regression. Journal of the American Statistical Association\/ 79 , 871--880

    work page 1984

  72. [72]

    Rousseeuw, P. J. (2026). Analyzing cellwise weighted data. Econometrics and Statistics\/ 38 , 31--41, DOI https://doi.org/10.1016/j.ecosta.2023.01.007

    work page doi:10.1016/j.ecosta.2023.01.007 2026

  73. [73]

    Rousseeuw, P. J. and C. Croux (1993). Alternatives to the median absolute deviation. Journal of the American Statistical Association\/ 88 , 1273--1283

    work page 1993

  74. [74]

    Rousseeuw, P. J. and W. Van den Bossche (2018). Detecting deviating data cells. Technometrics\/ 60 , 135--145

    work page 2018

  75. [75]

    Santos, F. (2020). Modern methods for old data: An overview of some robust methods for outliers detection with applications in osteology. Journal of Archaeological Science: Reports\/ 32 , 102423

    work page 2020

  76. [76]

    Saraceno, G. and C. Agostinelli (2021). Robust multivariate estimation based on statistical depth filters. TEST\/ 30 , 1--25

    work page 2021

  77. [77]

    Alqallaf, and C

    Saraceno, G., F. Alqallaf, and C. Agostinelli (2021). A R obust S eemingly U nrelated R egressions F or R ow- W ise A nd C ell- W ise C ontamination. arXiv preprint arXiv:2107.00975\/

    work page arXiv 2021

  78. [78]

    She, Y. and A. B. Owen (2011). Outlier detection using nonconvex penalized regression. Journal of the American Statistical Association\/ 106\/ (494), 626--639

    work page 2011

  79. [79]

    a tzungen: infinitesimale O ptimalit \

    Stahel, W. A. (1981). Robuste S ch \"a tzungen: infinitesimale O ptimalit \"a t und S ch \"a tzungen von K ovarianzmatrizen . Ph.\ D. thesis, Fachgruppe f\"ur Statistik

    work page 1981

  80. [80]

    Alfons, J

    S tefelov \'a , N., A. Alfons, J. Palarea-Albaladejo, P. Filzmoser, and K. Hron (2021). Robust regression with compositional covariates including cellwise outliers. Advances in Data Analysis and Classification\/ 15 , 869–909

    work page 2021

Showing first 80 references.