pith. sign in

arxiv: 2606.25155 · v1 · pith:T3V6GCTUnew · submitted 2026-06-23 · ⚛️ physics.soc-ph · physics.data-an

Wasserstein recurrence networks for multiscale time series pattern analysis

B\'eatrice D\'esy , Nicholas R. Golledge , Hana Ishii , Markus Luczak-Roesch This is my paper

Pith reviewed 2026-06-25 21:48 UTC · model grok-4.3

classification ⚛️ physics.soc-ph physics.data-an
keywords recurrence analysisWasserstein distancepaleoclimate time seriesmultiscale patternsBrownian motionthreshold selectionice core records
0
0 comments X

The pith

The 1-Wasserstein distance between time series patterns is scale-invariant under Brownian motion, supplying a threshold for identifying recurrences at any scale.

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

The paper develops a recurrence detection method for multiscale time series that have irregular sampling and abrupt shifts, such as paleoclimate archives. It applies a transform of the 1-Wasserstein distance so that repeating patterns appear as local minima below a threshold taken from the probability of equally similar patterns in one-dimensional Brownian motion. Numerical checks show that the distance distributions remain scale-invariant in Brownian motion, removing the need for arbitrary cutoffs that vary with resolution. A reader would care because the approach can flag climatic events whose durations range over two orders of magnitude within a single record.

Core claim

By demonstrating numerically the scale invariance of 1-Wasserstein distance distributions between patterns drawn from one-dimensional Brownian motion, the framework defines recurrences at any time scale as local minima of the distance that fall below the probability threshold given by that null model; non-overlapping recurrences can then be aggregated to produce a set of distinct multiscale recurring events, as illustrated on ice-rafted debris and ice-core records.

What carries the argument

The 1-Wasserstein distance applied to time-series segments via a mathematical transform, with recurrence thresholds taken from the scale-invariant distribution of distances in one-dimensional Brownian motion.

If this is right

  • Recurrences can be identified consistently across scales from seasonal to multi-millennial within the same record.
  • Non-overlapping aggregation produces a distinct collection of multiscale recurring events.
  • The method applies directly to records with abrupt changes and irregular temporal resolution.
  • Detected events in ice-core and debris records exhibit durations spanning more than two orders of magnitude.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same Brownian-derived threshold rule could be tested on synthetic series that embed known repeating patterns at multiple scales.
  • If the identified recurrences align with independently dated climate events, the framework would supply an objective way to compare recurrence statistics across different archives.
  • Extension to multivariate or higher-dimensional series would require checking whether the scale-invariance property survives the change in distance definition.

Load-bearing premise

That the statistical properties of one-dimensional Brownian motion supply an appropriate null model and threshold for determining what counts as a recurrence in real paleoclimate time series.

What would settle it

Empirical 1-Wasserstein distance distributions computed on paleoclimate records that deviate markedly from the scale-invariant form predicted by Brownian motion would falsify the threshold rule.

Figures

Figures reproduced from arXiv: 2606.25155 by B\'eatrice D\'esy, Hana Ishii, Markus Luczak-Roesch, Nicholas R. Golledge.

Figure 1
Figure 1. Figure 1: FIG. 1. Principle of the Wasserstein distance between dis [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. Definition of pattern recurrences using local minima [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. (a) An example of Brownian motion trajectory and various pattern durations [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4. From detection at multiple time scales [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5. Identifying similar patterns in different IRD records. (a) Southern hemisphere (orange) and (b) Northern hemisphere [PITH_FULL_IMAGE:figures/full_fig_p009_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: FIG. 6. Multiscale recurrence network for EDC ice core [PITH_FULL_IMAGE:figures/full_fig_p010_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: FIG. 7. Properties of the EDC multiscale recurrence network. [PITH_FULL_IMAGE:figures/full_fig_p011_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: FIG. 8. Cumulative distribution functions of the Wasser [PITH_FULL_IMAGE:figures/full_fig_p012_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: FIG. 9. (a) Euclidean distance distributions between patterns in Brownian motion at different time scales. (b) Euclidean [PITH_FULL_IMAGE:figures/full_fig_p013_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: FIG. 10. Number of edges in EDC multiscale recurrence [PITH_FULL_IMAGE:figures/full_fig_p013_10.png] view at source ↗
read the original abstract

Time series data are often generated by systems which operate on multiple temporal scales, of which Earth's climate system is a paramount example. Variations in global climate are recorded in paleo-environmental archives as temporal patterns across a wide range of time scales, from seasonal or decadal to multi-millennial. In this context, recurrence analysis, where repeating patterns are identified in time series, is limited by the underlying properties of the distance function used and of the time series data themselves, especially in terms of temporal resolution and scale dependence. In this paper, we present a novel recurrence analysis framework designed for multiscale time series data with abrupt changes and irregular temporal resolution as found in paleoclimate records. We introduce a simple mathematical transform to use the $1-$Wasserstein distance for recurring pattern detection in time series. The scale invariance of $1-$Wasserstein distance distributions between patterns in Brownian motion is demonstrated numerically, which provides a principled threshold choice for recurrences. At any time scale, recurrences are defined as local minima of the distance, granted that they are below a threshold given by the probability of encountering patterns at least as similar in one-dimensional Brownian motion. Recurrences can be further combined according to a non-overlapping condition to yield a distinct set of multiscale recurring events. We provide examples of climatic applications from ice-rafted debris and ice core records, where detected recurrences have durations spanning over two orders of magnitude. Our method extends recurrence analysis to more complex time series data and provides new avenues for statistical identification and analyses of recurring events at multiple temporal scales.

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

1 major / 2 minor

Summary. The manuscript introduces a recurrence analysis framework for multiscale time series with irregular resolution and abrupt changes, using a mathematical transform of the 1-Wasserstein distance. Recurrences are defined as local minima of this distance that fall below a threshold derived from the tail probability under one-dimensional Brownian motion, justified by a numerical demonstration of scale invariance in the W1 distance distributions for BM patterns. Recurrences are then combined under a non-overlapping condition to identify multiscale events. Applications are shown on ice-rafted debris and ice core records, with detected recurrences spanning over two orders of magnitude in duration.

Significance. If the BM-derived threshold is shown to be appropriate, the approach supplies a scale-invariant, numerically grounded method for recurrence detection in non-stationary paleoclimate series, extending recurrence analysis beyond standard distance functions and regular sampling. The explicit numerical demonstration of scale invariance under BM is a positive feature that supports the parameter-light threshold choice.

major comments (1)
  1. [Abstract (threshold definition)] Abstract and threshold construction: the claim that the threshold is 'principled' depends on 1D Brownian motion supplying a suitable null model whose distance statistics match those of the target paleoclimate series. Paleoclimate records typically contain trends, regime shifts, and long-memory correlations absent from standard BM; without explicit checks (e.g., comparison of empirical W1 distributions or surrogate tests preserving the series' autocorrelation structure), the threshold lacks calibrated meaning for the claimed applications.
minor comments (2)
  1. [Abstract] The 'simple mathematical transform' that enables use of the 1-Wasserstein distance on time series is referenced but not shown; including the explicit mapping or pseudocode would aid reproducibility.
  2. The title refers to 'recurrence networks' yet the abstract and central construction emphasize recurrence detection and combination; a short clarification of how the detected events form or are analyzed as a network would improve consistency.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive feedback. We address the single major comment below.

read point-by-point responses

  1. Referee: [Abstract (threshold definition)] Abstract and threshold construction: the claim that the threshold is 'principled' depends on 1D Brownian motion supplying a suitable null model whose distance statistics match those of the target paleoclimate series. Paleoclimate records typically contain trends, regime shifts, and long-memory correlations absent from standard BM; without explicit checks (e.g., comparison of empirical W1 distributions or surrogate tests preserving the series' autocorrelation structure), the threshold lacks calibrated meaning for the claimed applications.

    Authors: We agree that standard Brownian motion does not reproduce trends, regime shifts, or long-memory correlations typical of paleoclimate records, and that the manuscript does not include surrogate tests that preserve autocorrelation structure or direct comparisons of empirical W1 distributions. The BM reference is used solely to exploit the numerically demonstrated scale invariance of the W1 distance tail probabilities, thereby supplying a single, scale-consistent threshold without additional parameters. This is a modeling choice for threshold construction rather than a claim that BM statistics match the data. In revision we will (i) rephrase the abstract and methods to avoid the word 'principled' and instead describe the threshold as 'scale-invariant under a Brownian-motion reference,' (ii) add an explicit limitations paragraph noting the mismatch with paleoclimate autocorrelation, and (iii) outline how surrogate-based calibration could be performed in future applications. No new numerical checks will be added at this stage. revision: partial

Circularity Check

0 steps flagged

No significant circularity; threshold from external 1D Brownian motion null model

full rationale

The paper's central construction defines recurrences as local minima of 1-Wasserstein distances below a threshold taken from the tail probability under 1D Brownian motion, after numerically demonstrating scale invariance of the distance distribution on BM patterns. This uses an external statistical model as null benchmark rather than fitting parameters to the target paleoclimate series or reducing the result to the input data by construction. No self-citation load-bearing steps, uniqueness theorems imported from authors, ansatzes smuggled via citation, or renaming of known results are present. The derivation remains self-contained against the external BM benchmark; any mismatch with paleoclimate correlation structure is a question of model appropriateness, not circularity.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on a numerical demonstration of scale invariance whose details are not supplied and on the assumption that Brownian motion statistics transfer to paleoclimate series.

axioms (1)
  • domain assumption 1-Wasserstein distance distributions between patterns are scale-invariant in one-dimensional Brownian motion
    Invoked to justify the threshold choice for recurrences at any time scale

pith-pipeline@v0.9.1-grok · 5828 in / 1292 out tokens · 19807 ms · 2026-06-25T21:48:35.454774+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

54 extracted references · 2 canonical work pages

  1. [1]

    as unlikely as within purely uncorrelated noise

    for 0< t 1 < t2 < . . . < t n−1 < t n, the increments B(t1), B(t2)−B(t 1), . . . , B(tn)−B(t n−1) are independent and normally distributed, with means 0 and variancest 1, t2 −t 1, . . . , tn −t n−1. It follows from the latter property that it is scale in- variant,i.e.a −1B(a2t) is also a Brownian motion for alla >0. This property and its analytical simpli...

  2. [2]

    Paleoclimate Perspectives on Contemporary Climate Change,

    Sandy P. Harrison, Patrick J. Bartlein, Esmeralda Cruz- Silva, Olivia Haas, Stephen T. Jackson, Nikita Kaushal, Mengmeng Liu, Donatella Magri, Dominic T. Robson, Guido Vettoretti, and I. Colin Prentice, “Paleoclimate Perspectives on Contemporary Climate Change,” Annual Review of Environment and Resources50, 67–95 (2025)

    2025

  3. [3]

    What can Palaeoclimate Mod- elling do for you?

    A. M. Haywood, P. J. Valdes, T. Aze, N. Barlow, A. Burke, A. M. Dolan, A. S. von der Heydt, D. J. Hill, S. S. R. Jamieson, B. L. Otto-Bliesner, U. Salzmann, E. Saupe, and J. Voss, “What can Palaeoclimate Mod- elling do for you?” Earth Systems and Environment3, 1–18 (2019)

    2019

  4. [4]

    Paleocli- mate data provide constraints on climate models’ large- scale response to past CO2 changes,

    Daniel J. Lunt, Bette L. Otto-Bliesner, Chris Brier- ley, Alan Haywood, Gordon N. Inglis, Kenji Izumi, Masa Kageyama, Darrell Kaufman, Thorsten Mauritsen, Erin L. McClymont, Ulrich Salzmann, Sebastian Steinig, Jessica E. Tierney, Anni Zhao, and Jiang Zhu, “Paleocli- mate data provide constraints on climate models’ large- scale response to past CO2 changes...

    2024

  5. [5]

    Ad- vances in Paleoclimate Data Assimilation,

    Jessica E. Tierney, Emily J. Judd, Matthew B. Osman, Jonathan M. King, Olivia J. Truax, Nathan J. Steiger, Daniel E. Amrhein, and Kevin J. Anchukaitis, “Ad- vances in Paleoclimate Data Assimilation,” Annual Re- view of Earth and Planetary Sciences53, 625–650 (2025)

    2025

  6. [6]

    Problems with the current prac- tice of spectral analysis in cyclostratigraphy: Avoiding false detection of regular cyclicity,

    Graham P. Weedon, “Problems with the current prac- tice of spectral analysis in cyclostratigraphy: Avoiding false detection of regular cyclicity,” Earth-Science Re- views235, 104261 (2022)

    2022

  7. [7]

    Cyclostratigraphy and the problem of astrochronologic testing,

    Stephen R. Meyers, “Cyclostratigraphy and the problem of astrochronologic testing,” Earth-Science Reviews190, 190–223 (2019)

    2019

  8. [8]

    Distinct roles for precession, obliquity, and eccentricity in Pleistocene 100-kyr glacial cycles,

    Stephen Barker, Lorraine E. Lisiecki, Gregor Knorr, So- phie Nuber, and Polychronis C. Tzedakis, “Distinct roles for precession, obliquity, and eccentricity in Pleistocene 100-kyr glacial cycles,” Science387, eadp3491 (2025)

    2025

  9. [9]

    An astronomi- cally dated record of Earth’s climate and its predictabil- ity over the last 66 million years,

    Thomas Westerhold, Norbert Marwan, Anna Joy Drury, Diederik Liebrand, Claudia Agnini, Eleni Anagnos- tou, James S. K. Barnet, Steven M. Bohaty, David De Vleeschouwer, Fabio Florindo, Thomas Frederichs, David A. Hodell, Ann E. Holbourn, Dick Kroon, Vittoria Lauretano, Kate Littler, Lucas J. Lourens, Mitchell Lyle, Heiko P¨ alike, Ursula R¨ ohl, Jun Tian, R...

    2020

  10. [10]

    A New Recurrence-Network-Based Time Series Analysis Approach for Characterizing System Dynamics,

    Guangyu Yang, Daolin Xu, and Haicheng Zhang, “A New Recurrence-Network-Based Time Series Analysis Approach for Characterizing System Dynamics,” En- tropy21, 45 (2019)

    2019

  11. [11]

    Varley and Olaf Sporns, “Network Anal- 13 FIG

    Thomas F. Varley and Olaf Sporns, “Network Anal- 13 FIG. 9. (a) Euclidean distance distributions between patterns in Brownian motion at different time scales. (b) Euclidean distance distributions between patterns in Brownian motion when patterns are normalised using Eqs. (4-6) before computing the standard Euclidean distance. In both cases, the distributi...

    work page doi:10.3389/fnins.2021.787068 2022

  12. [12]

    Identification of dynam- ical transitions in marine palaeoclimate records by re- currence network analysis,

    J. F. Donges, R. V. Donner, K. Rehfeld, N. Marwan, M. H. Trauth, and J. Kurths, “Identification of dynam- ical transitions in marine palaeoclimate records by re- currence network analysis,” Nonlinear Processes in Geo- physics18, 545–562 (2011)

    2011

  13. [13]

    Complex network approach for recurrence analysis of time series,

    Norbert Marwan, Jonathan F. Donges, Yong Zou, Reik V. Donner, and J¨ urgen Kurths, “Complex network approach for recurrence analysis of time series,” Physics Letters A373, 4246–4254 (2009)

    2009

  14. [14]

    Recurrence Plots of Dynamical Systems,

    J.-P. Eckmann, S. Oliffson Kamphorst, and D. Ruelle, “Recurrence Plots of Dynamical Systems,” Europhysics Letters4, 973 (1987)

    1987

  15. [15]

    Recurrence plots for the analysis of com- plex systems,

    Norbert Marwan, M. Carmen Romano, Marco Thiel, and J¨ urgen Kurths, “Recurrence plots for the analysis of com- plex systems,” Physics Reports438, 237–329 (2007)

    2007

  16. [16]

    The Ap- plication of Recurrence Quantification Analysis in Detec- tion of Abrupt Climate Change,

    Wen Zhang, Guoling Feng, and Qunqun Liu, “The Ap- plication of Recurrence Quantification Analysis in Detec- tion of Abrupt Climate Change,” Discrete Dynamics in Nature and Society2016, e2689429 (2016)

    2016

  17. [17]

    A complex network approach to study the extreme precip- itation patterns in a river basin,

    Ankit Agarwal, Ravi Kumar Guntu, Abhirup Banerjee, Mayuri Ashokrao Gadhawe, and Norbert Marwan, “A complex network approach to study the extreme precip- itation patterns in a river basin,” Chaos: An Interdisci- plinary Journal of Nonlinear Science32, 013113 (2022)

    2022

  18. [18]

    Nonlinear detection of paleoclimate- variability transitions possibly related to human evolu- tion,

    Jonathan F. Donges, Reik V. Donner, Martin H. Trauth, Norbert Marwan, Hans-Joachim Schellnhuber, and J¨ urgen Kurths, “Nonlinear detection of paleoclimate- variability transitions possibly related to human evolu- tion,” Proceedings of the National Academy of Sciences 108, 20422–20427 (2011)

    2011

  19. [19]

    Classifying past climate change in the Chew Bahir basin, southern Ethiopia, using recurrence quantification analysis,

    Martin H. Trauth, Asfawossen Asrat, Walter Duesing, Verena Foerster, K. Hauke Kraemer, Norbert Marwan, Mark A. Maslin, and Frank Schaebitz, “Classifying past climate change in the Chew Bahir basin, southern Ethiopia, using recurrence quantification analysis,” Cli- mate Dynamics53, 2557–2572 (2019)

    2019

  20. [20]

    High predictability potential of highly synchro- nized widespread floods in monsoon regions,

    Jianxin Zhang, Kai Liu, Ming Wang, Kaiwen Li, Fenying Cai, Josef Ludescher, J¨ urgen Kurths, and Norbert Mar- wan, “High predictability potential of highly synchro- nized widespread floods in monsoon regions,” Journal of 14 Hydrology668, 135006 (2026)

    2026

  21. [21]

    How to avoid potential pitfalls in re- currence plot based data analysis,

    Norbert Marwan, “How to avoid potential pitfalls in re- currence plot based data analysis,” International Journal of Bifurcation and Chaos21, 1003–1017 (2011)

    2011

  22. [22]

    Ambiguities in recurrence-based com- plex network representations of time series,

    Reik V. Donner, “Ambiguities in recurrence-based com- plex network representations of time series,” Physical Re- view E81(2010), 10.1103/PhysRevE.81.015101

    work page doi:10.1103/physreve.81.015101 2010

  23. [23]

    Selection of recurrence threshold for signal detection,

    S. Schinkel, O. Dimigen, and N. Marwan, “Selection of recurrence threshold for signal detection,” The European Physical Journal Special Topics164, 45–53 (2008)

    2008

  24. [24]

    Recurrence threshold selection for ob- taining robust recurrence characteristics in different em- bedding dimensions,

    K. Hauke Kraemer, Reik V. Donner, Jobst Heitzig, and Norbert Marwan, “Recurrence threshold selection for ob- taining robust recurrence characteristics in different em- bedding dimensions,” Chaos: An Interdisciplinary Jour- nal of Nonlinear Science28, 085720 (2018)

    2018

  25. [25]

    Recurrence plot analysis of irregularly sampled data,

    Ibrahim Ozken, Deniz Eroglu, Sebastian F. M. Breit- enbach, Norbert Marwan, Liangcheng Tan, Ugur Tir- nakli, and J¨ urgen Kurths, “Recurrence plot analysis of irregularly sampled data,” Physical Review E98, 052215 (2018)

    2018

  26. [26]

    Local minima-based recurrence plots for continuous dynamical systems,

    Aaron P. Schultz, Yong Zou, Norbert Marwan, and Michael T. Turvey, “Local minima-based recurrence plots for continuous dynamical systems,” International Journal of Bifurcation and Chaos21, 1065–1075 (2011)

    2011

  27. [27]

    Short- and long-term variabil- ity of the Antarctic and Greenland ice sheets,

    Edward Hanna, D´ aniel Top´ al, Jason E. Box, Sammie Buzzard, Frazer D. W. Christie, Christine Hvidberg, Mathieu Morlighem, Laura De Santis, Alessandro Sil- vano, Florence Colleoni, Ingo Sasgen, Alison F. Ban- well, Michiel R. van den Broeke, Robert DeConto, Jan De Rydt, Heiko Goelzer, Alexandra Gossart, G. Hilmar Gudmundsson, Katrin Lindb¨ ack, Bertie Mi...

    2024

  28. [28]

    When Resources Collide: Towards a Theory of Coincidence in Information Spaces,

    Markus Luczak-Roesch, Ramine Tinati, and Nigel Shad- bolt, “When Resources Collide: Towards a Theory of Coincidence in Information Spaces,” inProceedings of the 24th International Conference on World Wide Web, WWW ’15 Companion (Association for Computing Ma- chinery, New York, NY, USA, 2015) pp. 1137–1142

    2015

  29. [29]

    What an Entangled Web We Weave: An Information-centric Approach to Time- evolving Socio-technical Systems,

    Markus Luczak-Roesch, Kieron O’Hara, Jesse David Din- neen, and Ramine Tinati, “What an Entangled Web We Weave: An Information-centric Approach to Time- evolving Socio-technical Systems,” Minds and Machines 28, 709–733 (2018)

    2018

  30. [30]

    Statistical As- pects of Wasserstein Distances,

    Victor M. Panaretos and Yoav Zemel, “Statistical As- pects of Wasserstein Distances,” Annual Review of Statistics and Its Application6, 405–431 (2019)

    2019

  31. [31]

    Berger, B

    C´ edric Villani,Optimal Transport, edited by M. Berger, B. Eckmann, P. De La Harpe, F. Hirzebruch, N. Hitchin, L. H¨ ormander, A. Kupiainen, G. Lebeau, M. Ratner, D. Serre, Ya. G. Sinai, N. J. A. Sloane, A. M. Vershik, and M. Waldschmidt, Grundlehren Der Mathematischen Wissenschaften, Vol. 338 (Springer, Berlin, Heidelberg, 2009)

    2009

  32. [32]

    The Earth Mover’s Distance as a Metric for Image Re- trieval,

    Yossi Rubner, Carlo Tomasi, and Leonidas J. Guibas, “The Earth Mover’s Distance as a Metric for Image Re- trieval,” International Journal of Computer Vision40, 99–121 (2000)

    2000

  33. [33]

    A linear transportation Lp distance for pattern recognition,

    Oliver M. Crook, Mihai Cucuringu, Tim Hurst, Carola- Bibiane Sch¨ onlieb, Matthew Thorpe, and Konstanti- nos C. Zygalakis, “A linear transportation Lp distance for pattern recognition,” Pattern Recognition147, 110080 (2024)

    2024

  34. [34]

    Wasser- stein distances in the analysis of time series and dynami- cal systems,

    Michael Muskulus and Sjoerd Verduyn-Lunel, “Wasser- stein distances in the analysis of time series and dynami- cal systems,” Physica D: Nonlinear Phenomena240, 45– 58 (2011)

    2011

  35. [35]

    The Wasserstein-Fourier Distance for Stationary Time Se- ries,

    Elsa Cazelles, Arnaud Robert, and Felipe Tobar, “The Wasserstein-Fourier Distance for Stationary Time Se- ries,” IEEE Transactions on Signal Processing69, 709– 721 (2021)

    2021

  36. [36]

    Computational Op- timal Transport with Applications to Data Sciences,

    Peyr´ e Gabriel and Cuturi Marco, “Computational Op- timal Transport with Applications to Data Sciences,” Foundations and Trends in Machine Learning11, 355– 607 (2019)

    2019

  37. [37]

    Optimal Mass Transport: Signal processing and machine-learning ap- plications,

    Soheil Kolouri, Se Rim Park, Matthew Thorpe, De- jan Slepcev, and Gustavo K. Rohde, “Optimal Mass Transport: Signal processing and machine-learning ap- plications,” IEEE Signal Processing Magazine34, 43–59 (2017)

    2017

  38. [38]

    Prescribing a System of Random Vari- ables by Conditional Distributions,

    R. L. Dobrushin, “Prescribing a System of Random Vari- ables by Conditional Distributions,” Theory of Probabil- ity & Its Applications15, 458–486 (1970)

    1970

  39. [39]

    On Wasserstein Two Sample Testing and Related Fam- ilies of Nonparametric Tests,

    Aaditya Ramdas, Nicolas Garcia, and Marco Cuturi, “On Wasserstein Two Sample Testing and Related Fam- ilies of Nonparametric Tests,” (2015), arXiv:1509.02237 [math, stat]

    Pith/arXiv arXiv 2015

  40. [40]

    (Springer, New York, NY, 1983)

    David Freedman,Brownian Motion and Diffusion, 1st ed. (Springer, New York, NY, 1983)

    1983

  41. [41]

    P Value and the Theory of Hypothesis Test- ing: An Explanation for New Researchers,

    David Jean Biau, Brigitte M. Jolles, and Rapha¨ el Porcher, “P Value and the Theory of Hypothesis Test- ing: An Explanation for New Researchers,” Clinical Or- thopaedics and Related Research468, 885–892 (2010)

    2010

  42. [42]

    Code for analysis and figures for pa- per

    B´ eatrice D´ esy, “Code for analysis and figures for pa- per...” Zenodo (2026)

    2026

  43. [43]

    Scientific method: Statistical errors,

    Regina Nuzzo, “Scientific method: Statistical errors,” Na- ture506, 150–152 (2014)

    2014

  44. [44]

    Time series analysis via network science: Concepts and algorithms,

    Vanessa Freitas Silva, Maria Eduarda Silva, Pedro Ribeiro, and Fernando Silva, “Time series analysis via network science: Concepts and algorithms,” WIREs Data Mining and Knowledge Discovery11, e1404 (2021)

    2021

  45. [45]

    Detecting strange attractors in turbu- lence,

    Floris Takens, “Detecting strange attractors in turbu- lence,” inDynamical Systems and Turbulence, War- wick 1980, edited by David Rand and Lai-Sang Young (Springer, Berlin, Heidelberg, 1981) pp. 366–381

    1980

  46. [46]

    Persistent influence of obliquity on ice age terminations since the Middle Pleis- tocene transition,

    Petra Bajo, Russell N. Drysdale, Jon D. Woodhead, John C. Hellstrom, David Hodell, Patrizia Ferretti, An- tje H. L. Voelker, Giovanni Zanchetta, Teresa Rodrigues, Eric Wolff, Jonathan Tyler, Silvia Frisia, Christoph Sp¨ otl, and Anthony E. Fallick, “Persistent influence of obliquity on ice age terminations since the Middle Pleis- tocene transition,” Scien...

    2020

  47. [47]

    Millennial-scale variability in Antarctic ice- sheet discharge during the last deglaciation,

    M. E. Weber, P. U. Clark, G. Kuhn, A. Timmer- mann, D. Sprenk, R. Gladstone, X. Zhang, G. Lohmann, L. Menviel, M. O. Chikamoto, T. Friedrich, and C. Ohlwein, “Millennial-scale variability in Antarctic ice- sheet discharge during the last deglaciation,” Nature 510, 134–138 (2014)

    2014

  48. [48]

    A 3.3-Million-Year Record of Antarctic Iceberg Rafted Debris and Ice Sheet Evolution Quantified by Machine Learning,

    Claire E. Jasper, Blake Dyer, Brendan T. Reilly, Trevor Williams, Sidney Hemming, and Maureen E. Raymo, “A 3.3-Million-Year Record of Antarctic Iceberg Rafted Debris and Ice Sheet Evolution Quantified by Machine Learning,” Paleoceanography and Paleoclimatology39, e2024PA004897 (2024)

    2024

  49. [49]

    Early Interglacial Legacy of Deglacial Climate Instabil- ity,

    Stephen Barker, Gregor Knorr, Stephen Conn, Sian 15 Lordsmith, Dhobasheni Newman, and David Thornalley, “Early Interglacial Legacy of Deglacial Climate Instabil- ity,” Paleoceanography and Paleoclimatology34, 1455– 1475 (2019)

    2019

  50. [50]

    Orbital and Millennial Antarctic Climate Variability over the Past 800,000 Years,

    J. Jouzel, V. Masson-Delmotte, O. Cattani, G. Drey- fus, S. Falourd, G. Hoffmann, B. Minster, J. Nouet, J. M. Barnola, J. Chappellaz, H. Fischer, J. C. Gallet, S. Johnsen, M. Leuenberger, L. Loulergue, D. Luethi, H. Oerter, F. Parrenin, G. Raisbeck, D. Raynaud, A. Schilt, J. Schwander, E. Selmo, R. Souchez, R. Spahni, B. Stauffer, J. P. Steffensen, B. Ste...

    2007

  51. [51]

    Marie Bouchet, Ama¨ elle Landais, Antoine Grisart, Fr´ ed´ eric Parrenin, Fr´ ed´ eric Pri´ e, Roxanne Jacob, Elise Fourr´ e, Emilie Capron, Dominique Raynaud, Vladimir Ya Lipenkov, Marie-France Loutre, Thomas Extier, Anders Svensson, Etienne Legrain, Patricia Mar- tinerie, Markus Leuenberger, Wei Jiang, Florian Rit- terbusch, Zheng-Tian Lu, and Guo-Min Y...

    2023

  52. [52]

    Causal in- ference for time series,

    Jakob Runge, Andreas Gerhardus, Gherardo Varando, Veronika Eyring, and Gustau Camps-Valls, “Causal in- ference for time series,” Nature Reviews Earth & Envi- ronment4, 487–505 (2023)

    2023

  53. [53]

    Transfer entropy for finite data,

    Alec Kirkley, “Transfer entropy for finite data,” Physical Review E112, L052304 (2025)

    2025

  54. [54]

    A multiscale model for multivari- ate time series forecasting,

    Vahid Naghashi, Mounir Boukadoum, and Ab- doulaye Banire Diallo, “A multiscale model for multivari- ate time series forecasting,” Scientific Reports15, 1565 (2025)

    2025