pith. sign in

arxiv: 1906.08825 · v1 · pith:XNGTAYD4new · submitted 2019-06-20 · ⚛️ physics.data-an · cs.CG· nlin.CD

Horizon Visibility Graphs and Time Series Merge Trees are Dual

Colin Stephen This is my paper

Pith reviewed 2026-05-25 18:49 UTC · model grok-4.3

classification ⚛️ physics.data-an cs.CGnlin.CD
keywords horizon visibility graphshorizontal visibility graphsmerge treesfiltered complexespersistent homologytime series analysisalgebraic topologyduality
0
0 comments X

The pith

Horizon visibility graphs are dual to merge trees from filtered complexes of time series.

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

The paper establishes that horizon visibility graphs, introduced as a simple extension of horizontal visibility graphs, stand in duality with merge trees that arise naturally over a filtered complex associated to any time series. Horizontal visibility graphs appear as weak duals of the same trees. This supplies an algebraic topology setting in which general properties of visibility graphs can be proved and natural extensions developed. A reader would care because the link immediately makes tree reconstruction theorems, statistics of self-similar trees, and persistent homology tools available to time series analysis that already uses visibility graphs across many domains.

Core claim

Horizon visibility graphs are dual to merge trees arising naturally over a filtered complex associated to a time series, while horizontal visibility graphs are weak duals of these trees.

What carries the argument

Duality between horizon visibility graphs and merge trees over a filtered complex canonically associated to a time series.

If this is right

  • Tree-based reconstruction theorems apply directly to horizon visibility graphs.
  • Statistics of self-similar trees become usable for analyzing visibility graphs of time series.
  • Visibility graphs gain explicit relations to applied persistent homology.

Where Pith is reading between the lines

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

  • Algorithms that reconstruct or compare merge trees could be repurposed to compare or reconstruct time series via their horizon visibility graphs.
  • Persistent homology barcodes computed on the filtered complex might supply new invariants for classifying or forecasting time series.
  • The same filtered-complex construction might produce dual trees for other sequential data representations beyond visibility graphs.

Load-bearing premise

A filtered complex can be canonically associated to any time series such that the resulting merge tree is dual to the horizon visibility graph.

What would settle it

A concrete time series whose horizon visibility graph edges fail to match the branches of the dual of its associated merge tree.

Figures

Figures reproduced from arXiv: 1906.08825 by Colin Stephen.

Figure 1
Figure 1. Figure 1: FIG. 1. Weighted graph [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. A time series merge tree [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. The principal subtree Λ [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4. Merge trees and their horizontal and horizon visibility [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5. A time series [PITH_FULL_IMAGE:figures/full_fig_p005_5.png] view at source ↗
read the original abstract

In this paper we introduce the horizon visibility graph, a simple extension to the popular horizontal visibility graph representation of a time series, and show that it possesses a rigorous mathematical foundation in computational algebraic topology. This fills a longstanding gap in the literature on the horizontal visibility approach to nonlinear time series analysis which, despite a suite of successful applications across multiple domains, lacks a formal setting in which to prove general properties and develop natural extensions. The main finding is that horizon visibility graphs are dual to merge trees arising naturally over a filtered complex associated to a time series, while horizontal visibility graphs are weak duals of these trees. Immediate consequences include availability of tree-based reconstruction theorems, connections to results on the statistics of self-similar trees, and relations between visibility graphs and the emerging field of applied persistent homology.

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

0 major / 2 minor

Summary. The paper introduces the horizon visibility graph as an extension of the horizontal visibility graph for time series and establishes that horizon visibility graphs are dual to merge trees arising from a filtered complex associated to the time series (via a height function on a 1-complex whose edges encode consecutive points), while horizontal visibility graphs are weak duals obtained by relaxing strict height ordering. The argument uses only standard persistent-homology definitions, supplies an explicit construction, and handles the usual distinct-values convention by perturbation.

Significance. If the duality holds, the result supplies a rigorous topological foundation for visibility-graph methods in nonlinear time series analysis, enabling tree-based reconstruction theorems, connections to the statistics of self-similar trees, and direct relations to applied persistent homology. Credit is due for the self-contained construction, absence of hidden assumptions beyond the standard perturbation convention, and use of canonical persistent-homology objects rather than ad-hoc fittings.

minor comments (2)
  1. §2: the definition of the 1-complex could explicitly state the vertex set as the time-series indices before introducing the height function, to avoid any momentary ambiguity for readers unfamiliar with the construction.
  2. Figure 3: the caption should clarify whether the depicted merge tree is the full persistence diagram or the reduced tree after the duality correspondence is applied.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for their positive summary of the manuscript, recognition of its significance, and recommendation to accept.

Circularity Check

0 steps flagged

No significant circularity

full rationale

The paper supplies an explicit construction of a filtered complex from any time series (via a height function on the 1-complex of consecutive points) and proves a direct correspondence to the horizon visibility graph via standard persistent homology definitions. This is a mathematical duality result, not a prediction or fit. No load-bearing self-citations, fitted parameters renamed as predictions, or self-definitional steps appear in the provided abstract or skeptic analysis. The central claim is self-contained against external topological benchmarks and does not reduce to its inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

Abstract-only review limits visibility into parameters and axioms; the paper introduces one new object and relies on standard results from algebraic topology.

axioms (1)
  • standard math Standard results from computational algebraic topology and persistent homology apply to filtered complexes built from time series.
    Invoked to establish the duality (abstract, main finding paragraph).
invented entities (1)
  • horizon visibility graph no independent evidence
    purpose: Simple extension of the horizontal visibility graph that admits a duality to merge trees.
    Newly defined in the paper to fill the formal gap.

pith-pipeline@v0.9.0 · 5655 in / 1213 out tokens · 28219 ms · 2026-05-25T18:49:23.507593+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

33 extracted references · 33 canonical work pages · 1 internal anchor

  1. [1]

    Horizon Visibility Graphs and Time Series Merge Trees are Dual

    or (D)HVG of a time series τ = (x1,...,x N) is a network with nodes {1,...,N } and edges (i,j ) for each pair i < jsuch that i < k < jimplies xk < xi,xj. The undirected version omits the order of i and j. Despite its structural simplicity, this graph captures much of the geometry ofτ while remaining invariant under (positive) affine transformations. Exact a...

    work page internal anchor Pith review Pith/arXiv arXiv 1906

  2. [2]

    Beketayev, D

    K. Beketayev, D. Yeliussizov, D. Morozov, G. H. Weber, and B. Hamann. Measuring the distance between merge trees. Topological Methods in Data Analysis and Visual- ization III , pages 151–165, 2014

    work page 2014

  3. [3]

    Braga, L

    A. Braga, L. Alves, L. Costa, A. Ribeiro, M. de Jesus, A. Tateishi, and H. Ribeiro. Characterization of river flow fluctuations via horizontal visibility graphs. Physica A: Statistical Mechanics and its Applications , 444:1003– 1011, 2016

    work page 2016

  4. [4]

    J. Curry. The fiber of the persistence map for functions on the interval. Journal of Applied and Computational Topology, 2(3):301–321, Dec 2018

    work page 2018

  5. [5]

    Dershowitz and S

    N. Dershowitz and S. Zaks. Enumerations of ordered trees. Discrete Mathematics, 31(1):9–28, 1980

    work page 1980

  6. [6]

    M. Drmota. Random trees: an interplay between com- binatorics and probability . Springer Science & Business Media, 2009

    work page 2009

  7. [7]

    Edelsbrunner and J

    H. Edelsbrunner and J. Harer. Computational topology: An introduction. American Mathematical Society, 2010

    work page 2010

  8. [8]

    Emrani, T

    S. Emrani, T. Gentimis, and H. Krim. Persistent homol- ogy of delay embeddings and its application to wheeze detection. IEEE Signal Processing Letters , 21(4):459– 463, 2014

    work page 2014

  9. [9]

    S. N. Evans. Probability and Real Trees. Springer, 2006

    work page 2006

  10. [10]

    Flanagan and L

    R. Flanagan and L. Lacasa. Irreversibility of financial time series: a graph-theoretical approach. Physics Letters A, 380(20):1689–1697, 2016

    work page 2016

  11. [11]

    R. Ghrist. Barcodes: the persistent topology of data. Bulletin of the American Mathematical Society , 45(1):61– 75, 2008

    work page 2008

  12. [12]

    Gidea and Y

    M. Gidea and Y. Katz. Topological data analysis of fi- nancial time series: Landscapes of crashes. Physica A , 491:820–834, 2018

    work page 2018

  13. [13]

    J. L. Gross and T. W. Tucker. Topological graph theory. Dover Publications, 2001

    work page 2001

  14. [14]

    D. Gusfield. Algorithms on strings, trees, and sequences: computer science and computational biology . Cambridge University Press, 1997

    work page 1997

  15. [15]

    Gutin, T

    G. Gutin, T. Mansour, and S. Severini. A characteriza- tion of horizontal visibility graphs and combinatorics on words. Physica A, 390(12):2421–2428, 2011

    work page 2011

  16. [16]

    F. A. Khasawneh and E. Munch. Chatter detection in turning using persistent homology. Mechanical Systems and Signal Processing , 70:527–541, 2016

    work page 2016

  17. [17]

    F. A. Khasawneh and E. Munch. Topological data anal- ysis for true step detection in periodic piecewise con- stant signals. Proceedings of the Royal Society of Lon- don A: Mathematical, Physical and Engineering Sciences , 474(2218), 2018

    work page 2018

  18. [18]

    Kovchegov and I

    Y. Kovchegov and I. Zaliapin. Horton law in self-similar trees. Fractals, 24(02):1650017, 2016

    work page 2016

  19. [19]

    Luque and L

    B. Luque and L. Lacasa. Canonical horizontal visibil- ity graphs are uniquely determined by their degree se- quence. The European Physical Journal Special Topics , 226(3):383–389, 2017

    work page 2017

  20. [20]

    Luque, L

    B. Luque, L. Lacasa, F. Ballesteros, and J. Luque. Hor- izontal visibility graphs: Exact results for random time series. Physical Review E , 80(4):046103, 2009

    work page 2009

  21. [21]

    T. Madl. Network analysis of heart beat intervals using horizontal visibility graphs. In 2016 Computing in Car- diology Conference (CinC) , pages 733–736. IEEE, 2016

    work page 2016

  22. [22]

    Mittal and S

    K. Mittal and S. Gupta. Topological characterization and early detection of bifurcations and chaos in complex systems using persistent homology. Chaos, 27(5):051102, 2017

    work page 2017

  23. [23]

    A. M. Nu˜ nez, L. Lacasa, J. P. Gomez, and B. Luque. Visibility algorithms: A short review. In New Frontiers in Graph Theory . IntechOpen, 2012

    work page 2012

  24. [24]

    J. A. Perea, A. Deckard, S. B. Haase, and J. Harer. SW1PerS: Sliding windows and 1-persistence scoring; dis- covering periodicity in gene expression time series data. BMC bioinformatics , 16(1):257, 2015

    work page 2015

  25. [25]

    J. A. Perea and J. Harer. Sliding Windows and Persis- tence: An Application of Topological Methods to Signal Analysis. Foundations of Computational Mathematics , 15(3):799–838, 2014

    work page 2014

  26. [26]

    Saitou and M

    N. Saitou and M. Nei. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Molec- ular Biology and Evolution , 4(4):406–425, 1987

    work page 1987

  27. [27]

    Schleussner, D

    C.-F. Schleussner, D. Divine, J. F. Donges, A. Miettinen, and R. Donner. Indications for a north atlantic ocean circulation regime shift at the onset of the little ice age. Climate dynamics , 45(11-12):3623–3633, 2015

    work page 2015

  28. [28]

    J. R. Tempelman and F. A. Khasawneh. A look into chaos detection through topological data analysis. arXiv e-prints, arXiv:1902.05918, 2019

    work page arXiv 1902

  29. [29]

    Zaliapin and Y

    I. Zaliapin and Y. Kovchegov. Tokunaga and Horton self- similarity for level set trees of Markov chains. Chaos, Solitons & Fractals , 45(3):358–372, 2012

    work page 2012

  30. [30]

    Zhou, J.-L

    Y.-W. Zhou, J.-L. Liu, Z.-G. Yu, Z.-Q. Zhao, and V. Anh. Fractal and complex network analyses of protein molec- ular dynamics. Physica A: Statistical Mechanics and its Applications, 416:21–32, 2014

    work page 2014

  31. [31]

    G. Zhu, Y. Li, and P. P. Wen. Epileptic seizure detec- tion in eegs signals using a fast weighted horizontal vis- ibility algorithm. Computer methods and programs in biomedicine, 115(2):64–75, 2014

    work page 2014

  32. [32]

    A. J. Zomorodian. Topology for computing . Cambridge University Press, 2005

    work page 2005

  33. [33]

    Y. Zou, R. V. Donner, N. Marwan, J. F. Donges, and J. Kurths. Complex network approaches to nonlinear time series analysis. Physics Reports, 787:1 – 97, 2019

    work page 2019