arxiv: 2605.11264 · v1 · submitted 2026-05-11 · 🧮 math.PR

Recognition: no theorem link

Uniform sampling of multitype continuous-time Bienaym\'e-Galton-Watson trees

Juan Carlos Pardo, Osvaldo Angtuncio Hern\'andez, Simon C. Harris

Pith reviewed 2026-05-13 01:46 UTC · model grok-4.3

classification 🧮 math.PR

keywords multitype branching processescontinuous-time Bienaymé-Galton-Watsonuniform samplingspine decompositiongenealogyancestral structurechange of measuretype-dependent offspring

0 comments

The pith

Multitype continuous-time Bienaymé-Galton-Watson processes admit an explicit uniform sample of k individuals whose genealogy is given by a k-spine decomposition under a size-biased change of measure.

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

The paper constructs a way to pick k individuals uniformly at random from the population of a multitype continuous-time Bienaymé-Galton-Watson process at a fixed time T and then describe the full ancestral tree connecting them. The method works by first changing the probability measure so that k distinguished spines represent the sampled lineages while the overall population size becomes k-size biased and exponentially discounted. Under the new measure the process retains its branching Markov property, which makes it possible to write down the splitting times of the spines, the offspring distributions along them, and the type-dependent rules that govern how the ancestors of one type produce descendants of other types. These type-dependent features have no counterpart in the single-type case and therefore constitute the main new content of the work.

Core claim

We study the genealogy of a sample of k individuals taken uniformly without replacement from a continuous-time multitype Bienaymé-Galton-Watson process at fixed times. Our results require only that the process be non-simple and conservative and that every type has positive probability to eventually lead to all other types. A k-spine decomposition together with a suitable change of measure makes the distinguished spines form a uniform sample at time T while the population size is k-size biased and exponentially discounted. This construction preserves the branching Markov property and yields an explicit description of the genealogical tree at fixed times, including spine splitting times, type-

What carries the argument

The k-spine decomposition under a k-size-biased and exponentially discounted change of measure, which converts the uniform sampling problem into an explicit branching process whose spines carry the sampled lineages and whose type transitions encode the ancestral structure.

If this is right

The genealogical tree connecting any fixed number of uniformly sampled individuals at time T admits an explicit probabilistic description that incorporates type interactions.
Spine splitting times and offspring distributions can be written down in closed form and depend on the type of the parent spine.
The same construction supplies the starting point for studying limiting genealogies when the population size or time tends to infinity.
More general sampling schemes that are not uniform can be treated by further reweighting the same changed measure.

Where Pith is reading between the lines

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

The type-interaction rules identified here could be used to test whether observed population data are consistent with a multitype rather than a single-type branching model.
The construction suggests a natural way to define effective reproduction numbers that vary with type and could be estimated from sampled genealogies.
Because the change of measure is explicit, numerical simulation of the uniform sample becomes straightforward and can serve as a benchmark for approximation methods in larger multitype systems.

Load-bearing premise

The multitype branching process must be non-simple and conservative, and every type must have positive probability of eventually producing individuals of every other type.

What would settle it

Generate many realizations of a qualifying multitype process at time T, draw uniform k-samples directly from each realization, and check whether the empirical distribution of splitting times and type-dependent offspring counts matches the explicit laws derived from the k-spine construction.

read the original abstract

We study the genealogy of a sample of $k$ individuals taken uniformly without replacement from a continuous-time multitype Bienaym\'e--Galton--Watson process at fixed times. Our results are quite general, requiring only that the process be non-simple and conservative, and that every type has a positive probability to ``eventually lead to'' all other types within the population. The corresponding single-type case has recently been studied by Johnston (2019), Harris, Johnston, and Roberts (2020), and Harris, Johnston, and Pardo (2024). Our approach is based on a $k$-spine decomposition and a suitable change of measure under which the distinguished spines form a uniform sample at time $T$, while the population size is subject to $k$-size biasing and exponential discounting. This construction preserves a branching Markov property and yields an explicit description of the genealogical tree at fixed times. In particular, we characterise spine splitting times, offspring distributions, and type-dependent ancestral structures, revealing rich interactions between types that are absent in the single-type setting. The present results form the basis of a forthcoming series of papers in which limiting genealogical behaviour is analysed under various asymptotic regimes and more general sampling schemes by the authors, see Angtuncio et al. (2026b), (2026c) and (2026d).

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.

Desk Editor's Note private letter to a colleague

This paper extends single-type k-spine uniform sampling to multitype continuous-time branching processes with explicit type interaction characterizations.

read the letter

Hi, the punchline here is that this work generalizes the uniform sampling of genealogies in single-type continuous-time Bienaymé-Galton-Watson processes to the multitype case. Using a k-spine decomposition and a change of measure, they make the distinguished spines correspond to a uniform sample at fixed time T, with the population size-biased and exponentially discounted. This gives explicit descriptions of spine splitting times, offspring distributions, and type-dependent ancestral structures that capture interactions between types. They handle this under general conditions: the process is non-simple and conservative, and every type has positive probability to eventually lead to all others. The construction preserves the branching Markov property, which is important for the results to hold. The characterizations reveal type mixing effects that are new compared to the single-type papers by Johnston, Harris, and others. This sets up their planned series on limiting behavior and other sampling schemes. The soft spots are not central. The abstract presents a coherent plan, but the full paper would need to detail how the measure change accounts for type-specific branching rates without breaking the uniformity or the Markov property. The type connectivity assumption is clearly necessary for rich interactions but does narrow the scope to connected type systems. No evidence of circular reasoning or weak evidence in the high-level argument. This paper is for people working on branching process theory and applications in population genetics or structured models. Someone familiar with the single-type spine decompositions would get the most out of the multitype extensions and the explicit formulas. It is solid enough to deserve a serious referee who can check the derivations. I would recommend sending it for peer review.

Referee Report

2 major / 3 minor

Summary. The paper develops a k-spine decomposition and associated change of measure for the genealogy of a uniform sample of k individuals drawn without replacement from a continuous-time multitype Bienaymé-Galton-Watson process at a fixed time T. Under the assumptions that the process is non-simple and conservative and that every type has positive probability of eventually leading to every other type, the construction yields an explicit description of spine splitting times, offspring distributions, and type-dependent ancestral structures while preserving the branching Markov property; this extends the cited single-type results of Johnston (2019), Harris et al. (2020), and Harris et al. (2024).

Significance. If the technical details of the multitype decomposition and measure change are verified, the work supplies a foundational tool for sampling genealogies in multitype branching processes and isolates new type-interaction phenomena absent from the single-type setting. The explicit preservation of the branching Markov property and the positioning of the results as the basis for three forthcoming papers on limiting regimes constitute clear strengths in terms of both technical coherence and research-program continuity.

major comments (2)

[k-spine decomposition (likely §3)] The abstract states that the k-spine construction and change of measure preserve the branching Markov property for the multitype case, but the manuscript must supply the precise statement of this property (including the role of the type-transition kernel) and verify that the exponential discounting and k-size biasing do not introduce type-dependent dependencies that violate the Markov property; this verification is load-bearing for the claim that the distinguished spines form a uniform sample at time T.
[Assumptions and main theorems (likely §2)] The 'eventually lead to' condition on types is invoked to guarantee rich interactions, yet the paper should isolate exactly which theorems rely on it versus which hold under the weaker non-simple and conservative hypotheses alone; without this separation the scope of the main results remains unclear.

minor comments (3)

Notation for the multitype offspring distribution and the associated intensity matrix should be introduced with an explicit comparison to the single-type case to facilitate reading.
[Introduction] All references to the single-type papers should appear in the introduction with precise bibliographic details rather than parenthetical citations only.
Figure captions for any genealogical diagrams should explicitly label the distinguished spines and the type labels to match the textual description.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their positive evaluation and constructive suggestions, which will improve the clarity and precision of the manuscript. We address each major comment below, indicating the revisions we plan to implement.

read point-by-point responses

Referee: [k-spine decomposition (likely §3)] The abstract states that the k-spine construction and change of measure preserve the branching Markov property for the multitype case, but the manuscript must supply the precise statement of this property (including the role of the type-transition kernel) and verify that the exponential discounting and k-size biasing do not introduce type-dependent dependencies that violate the Markov property; this verification is load-bearing for the claim that the distinguished spines form a uniform sample at time T.

Authors: We agree that an explicit statement of the branching Markov property, including the precise role of the type-transition kernel, together with a verification that the change of measure preserves it, strengthens the foundation of the k-spine construction. In the revised manuscript we will add, in Section 3, a formal definition of the property for the multitype setting and a direct verification that neither the exponential discounting nor the k-size biasing introduces type-dependent dependencies that would violate the Markov property. This will support the uniformity claim for the distinguished spines at time T. revision: yes
Referee: [Assumptions and main theorems (likely §2)] The 'eventually lead to' condition on types is invoked to guarantee rich interactions, yet the paper should isolate exactly which theorems rely on it versus which hold under the weaker non-simple and conservative hypotheses alone; without this separation the scope of the main results remains unclear.

Authors: We accept that separating the assumptions will clarify the scope. While the 'eventually lead to' condition is used throughout to obtain the richest type-interaction phenomena, certain basic properties of the k-spine decomposition and the change of measure hold under the weaker non-simple and conservative hypotheses alone. In the revision we will add, in Section 2, an explicit breakdown stating which theorems require the full set of assumptions and which remain valid without the 'eventually lead to' condition. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation self-contained

full rationale

The paper extends single-type results via explicit k-spine decomposition and change-of-measure constructions that preserve the branching Markov property under stated assumptions (non-simple, conservative process with positive type-transition probabilities). These constructions are defined directly in the multitype setting and yield new characterizations of spine splitting times, offspring distributions, and type-dependent structures without reducing to fitted parameters, self-definitions, or load-bearing self-citations. Prior single-type citations provide context but are not invoked as uniqueness theorems or ansatzes that force the multitype claims; the central results follow from the multitype branching structure and are not equivalent to inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim rests on standard domain assumptions for multitype branching processes plus the technical conditions stated for the construction to work.

axioms (2)

domain assumption The process is non-simple and conservative
Explicitly required in the abstract for the results to hold.
domain assumption Every type has positive probability to eventually lead to all other types
Stated as necessary to ensure rich type interactions in the ancestral structures.

pith-pipeline@v0.9.0 · 5556 in / 1281 out tokens · 37161 ms · 2026-05-13T01:46:47.821295+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

23 extracted references · 23 canonical work pages

[1]

Athreya and Jyy-I Hong, Coalescence on supercritical B ellman- H arris branching processes , Taiwanese J

Krishna B. Athreya and Jyy-I Hong, Coalescence on supercritical B ellman- H arris branching processes , Taiwanese J. Math. 22 (2018), no. 1, 245--261. 3749363

work page 2018
[2]

Angtuncio, S.H

O. Angtuncio, S.H. Harris, and J.C. Pardo, Sampling schemes of multitype B ienaym\'e- G alton- W atson trees and limiting critical genealogies , Preprint (2026)

work page 2026
[3]

Angtuncio, S.H

O. Angtuncio, S.H. Harris, and J.C. Pardo, Limiting genealogies of (sub)super-critical multitype B ienaym\'e- G alton- W atson trees , Work in progress (2026)

work page 2026
[4]

Angtuncio, S.H

O. Angtuncio, S.H. Harris, and J.C. Pardo, Limiting genealogies of heavy-tailed multitype B ienaym\'e- G alton- W atson trees , Work in progress (2026)

work page 2026
[5]

K. B. Athreya and P. E. Ney, Branching processes, Dover Publications, Inc., Mineola, NY, 2004, Reprint of the 1972 original [Springer, New York; MR0373040]. 2047480

work page 2004
[6]

Florin Boenkost, F\'elix Foutel-Rodier, and Emmanuel Schertzer, The genealogy of nearly critical branching processes in varying environment, 2024

work page 2024
[7]

Theory Related Fields 117 (2000), no

Jean Bertoin and Jean-Fran c ois Le Gall, The B olthausen- S znitman coalescent and the genealogy of continuous-state branching processes , Probab. Theory Related Fields 117 (2000), no. 2, 249--266. 1771663

work page 2000
[8]

Theory Related Fields 187 (2023), no

F\'elix Foutel-Rodier and Emmanuel Schertzer, Convergence of genealogies through spinal decomposition with an application to population genetics, Probab. Theory Related Fields 187 (2023), no. 3-4, 697--751. 4664584

work page 2023
[9]

Jyy-I Hong and K. B. Athreya, Markov limit of line of decent types in a multitype supercritical branching process, Statist. Probab. Lett. 98 (2015), 54--58. 3306444

work page 2015
[10]

Simon Harris, Samuel G. G. Johnston, and Juan Carlos Pardo, Universality classes for the coalescent structure of heavy-tailed G alton- W atson trees , Ann. Probab. 52 (2024), no. 2, 387--433. 4718398

work page 2024
[11]

Harris, Samuel G

Simon C. Harris, Samuel G. G. Johnston, and Matthew I. Roberts, The coalescent structure of continuous-time G alton- W atson trees , Ann. Appl. Probab. 30 (2020), no. 3, 1368--1414. 4133376

work page 2020
[12]

Jyy-I Hong, Coalescence in B ellman- H arris and multi-type branching processes , ProQuest LLC, Ann Arbor, MI, 2011, Thesis (Ph.D.)--Iowa State University. 2942128

work page 2011
[13]

1, 65--78

, Coalescence on supercritical multi-type branching processes, Sankhya A 77 (2015), no. 1, 65--78. 3317481

work page 2015
[14]

, Coalescence on critical and subcritical multitype branching processes, J. Appl. Probab. 53 (2016), no. 3, 802--817. 3570095

work page 2016
[15]

Harris, Sandra Palau, and Juan Carlos Pardo, The coalescent structure of G alton- W atson trees in varying environments , Ann

Simon C. Harris, Sandra Palau, and Juan Carlos Pardo, The coalescent structure of G alton- W atson trees in varying environments , Ann. Appl. Probab. 34 (2024), no. 6, 5388--5425. 4840489

work page 2024
[16]

Theory Related Fields 134 (2006), no

Svante Janson, Limit theorems for triangular urn schemes, Probab. Theory Related Fields 134 (2006), no. 3, 417--452. 2226887

work page 2006
[17]

Samuel G. G. Johnston and Amaury Lambert, The coalescent structure of uniform and P oisson samples from multitype branching processes , Ann. Appl. Probab. 33 (2023), no. 6A, 4820--4857. 4674065

work page 2023
[18]

1626, Springer, Berlin, 1996, pp

Peter Jagers and Olle Nerman, The asymptotic composition of supercritical multi-type branching populations, S\' e minaire de P robabilit\' e s, XXX , Lecture Notes in Math., vol. 1626, Springer, Berlin, 1996, pp. 40--54. 1459475

work page 1996
[19]

Samuel G. G. Johnston, The genealogy of G alton- W atson trees , Electron. J. Probab. 24 (2019), Paper No. 94, 35. 4003147

work page 2019
[20]

Lea Popovic and Mariolys Rivas, The coalescent point process of multi-type branching trees, Stochastic Process. Appl. 124 (2014), no. 12, 4120--4148. 3264441

work page 2014
[21]

Savits, The explosion problem for branching M arkov process , Osaka Math

Thomas H. Savits, The explosion problem for branching M arkov process , Osaka Math. J. 6 (1969), 375--395. 282426

work page 1969
[22]

Seneta, Non-negative matrices, Halsted Press [John Wiley & Sons], New York, 1973, An introduction to theory and applications

E. Seneta, Non-negative matrices, Halsted Press [John Wiley & Sons], New York, 1973, An introduction to theory and applications. 389944

work page 1973
[23]

B. A. Sewastjanow, Verzweigungsprozesse, R. Oldenbourg Verlag, Munich-Vienna, 1975, \" U bersetzt aus dem Russischen von Walter Warmuth. 408019

work page 1975