Pith. sign in

REVIEW 3 major objections 7 minor 31 references

Speed of coming down from infinity for $\Lambda$-Fleming-Viot initial support

T0 review · 3 major / 7 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The paper finds the exact rate at which the support of a Λ-Fleming-Viot process collapses from unbounded to bounded, in terms of the dual coalescent's coming-down speed and the initial tail.

desk verdict First quantitative rates for LFV support compactness, but the scope is narrower than the abstract suggests: condition (3.3) fails for some super-polynomial speeds, and several proofs are omitted. read the letter →

arxiv 2506.07067 v1 pith:YN5LZGRM submitted 2025-06-08 math.PR

Huili Liu , Xiaowen Zhou This is my paper
classification math.PR MSC 60J6860G1760J9560G57
keywords Λ-Fleming-ViotprocesscomingdownfrominfinitycompactsupportpropertylookdownrepresentationΛ-coalescentextremalextremevaluetheorymodulusofcontinuity
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

Starting from an initial probability measure with unbounded support, the $\Lambda$-Fleming-Viot process with Brownian spatial motion almost surely acquires a compact support at every positive time — a 'coming down from infinity' of the support. The paper asks how fast this happens and answers it: the speed is governed by two ingredients, the speed function $v(t)$ of the dual $\Lambda$-coalescent coming down from infinity and the asymptotic inverse of the tail of the initial distribution. The main result is a distributional limit for the maximum position $\hat{M}(t)$ of individuals alive at time $t$: $P(\hat{M}(t) \le \bar{F}^{-1}(x/v(t))) \to e^{-x}$, under a mild regularity condition. For power-law tails this yields a Fréchet limit, and for exponential-type tails a Gumbel limit after affine normalization. A reader should care because this gives the first quantitative rates for the compact support phenomenon in this setting, connecting population genetics to extreme value theory.

What carries the argument

The argument runs on the lookdown representation, a countable particle system whose empirical measure is the $\Lambda$-Fleming-Viot process and whose ancestral partitions evolve as the dual $\Lambda$-coalescent. The two quantities that carry the proof are $N(t)$, the number of blocks in the coalescent at time $t$ (which satisfies $N(t)/v(t) \to 1$, so only $v(t)$ ancestors are visible at small times), and the extremal process $\hat{M}(t) = \sup_i X_i(t)$. A modulus-of-continuity bound (Proposition 2.5), proved under the integral-test Assumption 2.4, shows that each individual's displacement from its ancestor is at most $C(\delta) t^{\delta}$, so $\hat{M}(t)$ differs from the ancestral maximum $M(t) = \max_{i \le N(t)} X_i(0)$ by a negligible amount. Replacing $N(t)$ by $v(t)$ turns $M(t)$ into the maximum of $v(t)$ i.i.d. samples, and the asymptotic inverse function $\bar{F}^{-1}$ rescales that maximum into the standard exponential limit $e^{-x}$; the extra condition (3.3) guarantees that the $t^{\delta}$ displacements do not break this rescaling. Extreme value theory then converts the exponential limit into the explicit Fréchet and Gumbel limits for the two tail regimes.

What would settle it

Simulate the lookdown particle system for a Beta$(2-\beta,\beta)$ coalescent with $\beta \in (1,2)$ and power-law initial tails $\bar{F}(x) = x^{-r_2}$, and check whether $a(v(t)) \hat{M}(t)$ converges to the Fréchet law with tail $e^{-x^{-r_2}}$ as Theorem 3.7 predicts. Then repeat with a coalescent whose $\psi(q) \sim C q (\log q)^{1+\epsilon}$ for $\epsilon \in (0,1]$, for which Assumption 2.4 fails though coming down from infinity holds; if the same limit appears, the paper's integral-test condition is stronger than necessary.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is Theorem 3.4: if the dual $\Lambda$-coalescent comes down from infinity with speed $v$ defined by $\int_{v(t)}^{\infty} dq/\psi(q) = t$, and if the tail of the initial distribution is regular enough to satisfy condition (3.3), then the extremal process $\hat{M}(t)$ — the maximum of the locations of all individuals alive at time $t$, which equals the supremum of the support — satisfies $P(\hat{M}(t) \le \bar{F}^{-1}(x/v(t))) \to e^{-x}$ as $t \to 0+$. In looser terms, only about $v(t)$ of the infinitely many original types matter at time $t$, so the maximum of the support behaves like the maximum of $v(t)$ independent draws from the initial distribution, while the Brownian spatial motion contributes only a negligible displacement of order $t^{\delta}$. Under power-law tails $\bar{F}(x) \sim r_1 x^{-r_2} (\log x)^{r_3}$, this gives $a(v(t)) \hat{M}(t)$ converging to a Fréchet distribution with tail $e^{-x^{-r_2}}$; under exponential-type tails $\bar{F}(x) \sim r_4 x^{r_3} e^{-r_1 x^{r_2}}$, an affine renormalization converges to the Gumbel distribution. The speed of coming down from infinity for the support is thus $a^{-1}(v(t))$ in the slow regime and $(r_1^{-1} \log v(t))^{1/r_2}$ in the fast regime.

Load-bearing premise

The load-bearing premise is that the coalescent's speed function $v(t)$ decays quickly enough to satisfy the integral test $\int_{0+} v(t)^{1+\epsilon} e^{-t^{-\delta}} dt < \infty$, which is strictly stronger than merely coming down from infinity and excludes coalescents with $\psi(q) \sim C q (\log q)^{1+\epsilon}$ for $\epsilon \in (0,1]$.

Editorial extensions

If this is right

  • The speed of the support's collapse is $a^{-1}(v(t))$ for power-law tails and $(r_1^{-1} \log v(t))^{1/r_2}$ for exponential-type tails, so the more slowly the dual coalescent comes down from infinity, the more slowly the support shrinks.
  • $\hat{M}(t)$ and the support supremum coincide almost surely, so the distributional limits for $\hat{M}(t)$ apply directly to the random measure's compact support boundary.
  • Under regularly varying tails with common index $-r$, $\log \hat{M}(t)/\log v(t) \to r^{-1}$ in probability, giving a purely logarithmic version of the rate that requires only upper and lower regularly varying bounds.
  • For $r_2 > 1$ in the fast regime, $\hat{M}(t) - (r_1^{-1} \log v(t))^{1/r_2} \to 0$ in probability, so the leading deterministic location is identified up to first order; for all $r_2 > 0$ the ratio converges to 1.
  • The same convergence statements hold in $\mathbb{R}^d$ under tail behavior of the norm, as indicated by the paper's final section.

Reading between the lines

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

  • Where the dual coalescent comes down from infinity but violates the integral test ($\psi(q) \sim C q (\log q)^{1+\epsilon}$ with $\epsilon \in (0,1]$), the paper's rates are silent; a natural conjecture is that the support still collapses but at a rate reflecting a slower-than-power-law $v$, and the Brownian displacement may no longer be negligible relative to the ancestral maximum.
  • The mechanism 'only $v(t)$ ancestors matter, and Brownian motion is a $t^{\delta}$ correction' suggests a general recipe: for any spatial motion with a known modulus of continuity after the lookdown construction, the same argument should convert coalescent coming-down speeds into support coming-down speeds; for jump-type spatial motions, however, the support may propagate instantaneously instead o
  • Condition (3.3) is a local regularity requirement on the tail $\bar{F}$ at the moving point $\bar{F}^{-1}(x/v(t))$; checking it is equivalent to verifying that $\bar{F}$ is regularly varying enough at its own inverse, which connects the paper's hypotheses to standard domains-of-attraction conditions in extreme value theory.
  • In population-genetic terms, the result says that after an infinitesimal time the population's type space is effectively finite with size proportional to $v(t)$, and the rarest initially present type that survives is described by the extremal laws, so the paper quantifies the speed of loss of genetic diversity at the frontier of the type space.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 7 minor

Summary. The paper studies the Λ-Fleming-Viot process with Brownian spatial motion whose dual Λ-coalescent comes down from infinity. Using the lookdown representation, it establishes a modulus-of-continuity estimate under the integral test (Assumption 2.4) and then derives asymptotic results for the extremal process M̂(t), the maximum position of individuals alive at time t. The central result, Theorem 3.4, states that if the tail-stability condition (3.3) holds, then P(M̂(t) ≤ F̄^{-1}(x/v(t))) → e^{-x} as t → 0+, where v is the speed of coming down from infinity of the dual coalescent. This is specialized to power-law tails (Theorem 3.7, yielding a Fréchet limit) and exponential-type tails (Theorem 3.14, yielding a Gumbel limit), with additional convergence-in-probability results under weaker regularly varying tail bounds (Corollaries 3.11 and 3.21). The proof strategy combines the known speed theorem and moment estimates of Berestycki et al. [4] with the authors' earlier support and modulus results [26,27,28].

Significance. If the proof gaps are filled, this would be the first quantitative characterization of the speed at which the support of a Λ-Fleming-Viot process with Brownian motion becomes bounded, connecting the coalescent CDI speed to the extreme-value asymptotics of the initial measure. A strength of the paper is that the main hypothesis, condition (3.3), is an explicit and falsifiable separation-of-scales condition rather than a hidden consequence of the conclusion. The distinction between the ancestral maximum M(t) and the actual extremal process M̂(t) is handled through the modulus of continuity, which is the right conceptual route. However, several load-bearing proofs are currently omitted or only sketched, so the results as written are conditional on completing those arguments.

major comments (3)
  1. [Theorem 3.4] The proof of Theorem 3.4 stops with 'Since the proofs of the two are very similar, we omit the subsequent steps.' This omitted fluctuation step is exactly what converts the conditional estimates into the claimed limit: one must split on the event {N(t) ≥ (1-ε)v(t)} for the upper bound and {N(t) ≤ (1+ε)v(t)} for the lower bound, and handle the factor P(t < Δ) in the lower bound. As written, the central theorem of the paper is not fully proved; please include the complete argument.
  2. [Proposition 3.19] Proposition 3.19 is stated with no proof at all ('The proof of Proposition 3.19 follows a similar approach to that of Proposition 3.10. For brevity, we omit the details.'). This proposition is used in Corollary 3.21 to derive the convergence-in-probability results in the fast regime, so it is load-bearing. A full proof, or a precise reference giving the analogous argument, must be supplied.
  3. [Assumption 3.12 and Theorem 3.14] The fast-regime results are substantially more restrictive than Assumption 2.4 alone, and the paper should state this explicitly. For example, take v(t)=exp(t^{-1/2}) and F̄(x)=e^{-x^2}. Assumption 2.4 holds with any δo > 1/2, so Proposition 2.5 forces δ < 1/4. But for q = F̄^{-1}(x/v(t)) ~ t^{-1/4}, the error in condition (3.3) is of order v(t) · |F̄(q ± C t^δ) - x/v(t)| ≈ 2xC t^{δ-1/4}, which diverges for every admissible δ < 1/4. Thus condition (3.3) fails even though the speed v satisfies the paper's integrability assumption. This is not an internal contradiction because Assumption 3.12 excludes this case, but the current Remark 3.13 only gives a sufficient condition for Assumption 3.12 and does not flag the obstruction; the scope of the 'spatial motion is negligible' conclusion should be clarified.
minor comments (7)
  1. [Title and Section 1] The title contains a typo, 'SUPPOR T' for 'SUPPORT', and in Section 1 the phrase 'seminal wok' should read 'seminal work'.
  2. [Definition 3.1] In Definition 3.1, 'Not that the asymptotic inverse function F̄^{(-1)} is not unique' should read 'Note that'.
  3. [Corollary 3.20] In Corollary 3.20 the symbol ν(t) appears in the displayed limit; this should be the speed function v(t).
  4. [Proof of Corollary 3.21] In the proof of Corollary 3.21, the notation '¯d2(v(t))' appears where 'd2(v(t))' is intended.
  5. [Remark 3.22] The word 'Combing' in Remark 3.22 should be 'Combining'.
  6. [Abstract] The abstract says the support 'becomes finite as soon as t>0'; since the support is a subset of R, this should read 'compact' (or 'bounded').
  7. [Proposition 2.5] Proposition 2.5 is stated for arbitrary T > 0, but the proof in Appendix A is written for the time interval [0,1]. The extension to general T via Brownian scaling should be explicitly indicated.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the support-CDI speeds are derived from the published coalescent speed theorem, a modulus bound proved in the appendix, and explicit tail hypotheses; no fitted input is renamed as a prediction.

full rationale

The derivation chain is self-contained. The input ingredients are: (i) Berestycki et al. [4] gives the coalescent speed v via (2.2) and the laws of large numbers (2.3)-(2.4); (ii) Proposition 2.5, the modulus-of-continuity estimate, is proved in Appendix A under Assumption 2.4, so it is not imported as a black box; (iii) the tail classes are explicit hypotheses (3.4) and (3.13), and Assumption 3.12 is a stated separation-of-scales condition. Lemma 3.2 is a direct extreme-value calculation: with N(t)/v(t) -> 1 and Fbar(Fbar^{-1}(y)) ~ y, the maximum of N(t) i.i.d. variables at the quantile Fbar^{-1}(x/v(t)) converges to e^{-x}; no parameter is fitted. Theorem 3.4 transfers this to the extremal process Mhat using the modulus bound (3.1) and condition (3.3), which is an explicit tail-continuity hypothesis checked in each corollary rather than a consequence of the conclusion. The self-citations [26,27,28] supply earlier support identities and technical estimates (e.g., v(t) >= 2/t referenced from [28]); these are independent published results, and the central modulus bound is reproven here. Remark 2.7 and Assumption 3.12 are candid limitations: Assumption 2.4 is strictly stronger than CDI, and the fast regime requires (log v(t))^{0∨(1-1/r2)} t^δ -> 0. These restrict scope but do not make the argument circular. Proposition 3.19 is stated without proof, but it is the one-sided analogue of Proposition 3.10 and does not conceal a reduction of the theorem to itself.

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

The results are conditional on the CDI property, the additional integrability condition (2.5), the i.i.d. initial condition, and a tail-stability condition (3.3). These are stated hypotheses rather than hidden fits. The tail constants r1, r2, r3, r4 are properties of the initial distribution F, not fitted parameters. No new particles, forces, or entities are introduced.

assumptions (5)
  • domain assumption The dual Lambda-coalescent comes down from infinity, with speed v defined by (2.2).
    Section 3 assumes CDI throughout; the compact support property and the rates are only studied for CDI coalescents.
  • domain assumption Assumption 2.4: for some delta_o in (0,1) and epsilon_o > 0, integral_{0+} v(t)^{1+epsilon_o} e^{-t^{-delta_o}} dt < infinity.
    Required for the modulus of continuity bound in Proposition 2.5; excludes some CDI coalescents as noted in Remark 2.7.
  • domain assumption The initial locations X1(0), X2(0), ... are i.i.d. with common distribution F.
    The extreme value computations for M(t) rely on independence and identical distribution of the initial types.
  • domain assumption Condition (3.3): v(t) | Fbar(Fbar^{-1}(x/v(t)) +/- C(delta)t^delta) - x/v(t)| tends to 0.
    Theorem 3.4 is conditional on this tail-stability condition; the slow and fast regimes verify it.
  • domain assumption Support identification S(X(t)) = {X1(t), X2(t), ...} from Lemma 2.2 (prior work [27]).
    Used to equate the support supremum with the particle maximum Mhat(t).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Speed of coming down from infinity for $\Lambda$-Fleming-Viot initial support." pith.science (2026). https://pith.science/paper/YN5LZGRM

@misc{pith2026250607067,
  author       = {Pith},
  title        = {Pith review of: Speed of coming down from infinity for $\Lambda$-Fleming-Viot initial support},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YN5LZGRM}},
  note         = {Machine review of arXiv:2506.07067}
}
abstract

The $\Lambda$-Fleming-Viot process is a probability measure-valued process that is dual to a $\Lambda$-coalescent that allows multiple collisions. In this paper, we consider a class of $\Lambda$-Fleming-Viot processes with Brownian spatial motion and with associated $\Lambda$-coalescents that come down from infinity. Notably, these processes have the compact support property: the support of the process becomes finite as soon as $t>0$, even though the initial measure has unbounded support. We obtain asymptotic results characterizing the rates at which the initial supports become finite. The rates of coming down are expressed in terms of the asymptotic inverse function of the tail distribution of the initial measure and the speed function of coming down from infinity for the corresponding $\Lambda$-coalescent.

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

31 extracted references · 30 canonical work pages

  1. [4]

    The Λ-coalescent speed of coming down from infinity

    Julien Berestycki, Nathana¨ el Berestycki, and Vlada Limic. The Λ-coalescent speed of coming down from infinity. Ann. Probab., 38(1):207–233, 2010

    work page 2010

  2. [1]

    David J. Aldous. Deterministic and stochastic models for coalescence (aggregation and coagulation): a review of the mean-field theory for probabilists. Bernoulli, 5(1):3–48, 1999

    work page 1999

  3. [2]

    Diffusions from infinity

    Vincent Bansaye, Pierre Collet, Servet Martinez, Sylvie M´ el´ eard, and Jaime San Martin. Diffusions from infinity. Trans. Amer. Math. Soc. , 372(8):5781–5823, 2019

    work page 2019

  4. [3]

    Speed of coming down from infinity for birth-and- death processes

    Vincent Bansaye, Sylvie M´ el´ eard, and Mathieu Richard. Speed of coming down from infinity for birth-and- death processes. Adv. in Appl. Probab. , 48(4):1183–1210, 2016

    work page 2016

  5. [5]

    Small-time behavior of beta coalescents

    Julien Berestycki, Nathana¨ el Berestycki, and Jason Schweinsberg. Small-time behavior of beta coalescents. Ann. Inst. Henri Poincar´ e Probab. Stat., 44(2):214–238, 2008

    work page 2008

  6. [6]

    Random fragmentation and coagulation processes , volume 102 of Cambridge Studies in Ad- vanced Mathematics

    Jean Bertoin. Random fragmentation and coagulation processes , volume 102 of Cambridge Studies in Ad- vanced Mathematics. Cambridge University Press, Cambridge, 2006

    work page 2006

  7. [7]

    Stochastic flows associated to coalescent processes

    Jean Bertoin and Jean-Francois Le Gall. Stochastic flows associated to coalescent processes. III. Limit theorems. Illinois J. Math. , 50(1-4):147–181, 2006

    work page 2006

  8. [8]

    N. H. Bingham, C. M. Goldie, and J. L. Teugels. Regular variation, volume 27 of Encyclopedia of Mathe- matics and its Applications . Cambridge University Press, Cambridge, 1989

    work page 1989

Show all 31 references
  1. [9]

    Generalised stable Fleming-Viot processes as flickering random mea- sures

    Matthias Birkner and Jochen Blath. Generalised stable Fleming-Viot processes as flickering random mea- sures. Electron. J. Probab., 14:no. 84, 2418–2437, 2009

    2009

  2. [10]

    Measure-valued diffusions, general coalescents and population genetic inference

    Matthias Birkner and Jochen Blath. Measure-valued diffusions, general coalescents and population genetic inference. In Trends in stochastic analysis , volume 353 of London Math. Soc. Lecture Note Ser. , pages 329–363. Cambridge Univ. Press, Cambridge, 2009

    2009

  3. [11]

    A modified lookdown construction for the Xi-Fleming-Viot process with mutation and populations with recurrent bottlenecks

    Matthias Birkner, Jochen Blath, Martin M¨ ohle, Matthias Steinr¨ ucken, and Johanna Tams. A modified lookdown construction for the Xi-Fleming-Viot process with mutation and populations with recurrent bottlenecks. ALEA Lat. Am. J. Probab. Math. Stat. , 6:25–61, 2009

    2009

  4. [12]

    Measure-valued processes, self-similarity and flickering random measures

    Jochen Blath. Measure-valued processes, self-similarity and flickering random measures. In Fractal geometry and stochastics IV , volume 61 of Progr. Probab., pages 175–196. Birkh¨ auser Verlag, Basel, 2009

    2009

  5. [13]

    Donald A. Dawson. Measure-valued Markov processes. In ´Ecole d’ ´Et´ e de Probabilit´ es de Saint-Flour XXI— 1991, volume 1541 of Lecture Notes in Math. , pages 1–260. Springer, Berlin, 1993

    1991

  6. [14]

    Dawson and Kenneth J

    Donald A. Dawson and Kenneth J. Hochberg. Wandering random measures in the Fleming-Viot model. Ann. Probab., 10(3):554–580, 1982

    1982

  7. [15]

    Dawson and Vladimir Vinogradov

    Donald A. Dawson and Vladimir Vinogradov. Almost-sure path properties of (2 , d, β)-superprocesses. Sto- chastic Process. Appl., 51(2):221–258, 1994

    1994

  8. [16]

    Kolmogorov’s test for super-Brownian motion

    Jean-St´ ephane Dhersin and Jean-Fran¸ cois Le Gall. Kolmogorov’s test for super-Brownian motion. Ann. Probab., 26(3):1041–1056, 1998. 28 Speed of CDI forΛ-Fleming-Viot support

    1998

  9. [17]

    Peter Donnelly and Thomas G. Kurtz. A countable representation of the Fleming-Viot measure-valued diffusion. Ann. Probab., 24(2):698–742, 1996

    1996

  10. [18]

    Peter Donnelly and Thomas G. Kurtz. Genealogical processes for Fleming-Viot models with selection and recombination. Ann. Appl. Probab., 9(4):1091–1148, 1999

    1999

  11. [19]

    Peter Donnelly and Thomas G. Kurtz. Particle representations for measure-valued population models. Ann. Probab., 27(1):166–205, 1999

    1999

  12. [20]

    Some mathematical models from population genetics , volume 2012 of Lecture Notes in Mathematics

    Alison Etheridge. Some mathematical models from population genetics , volume 2012 of Lecture Notes in Mathematics. Springer, Heidelberg, 2011. Lectures from the 39th Probability Summer School held in Saint- Flour, 2009, ´Ecole d’ ´Et´ e de Probabilit´ es de Saint-Flour. [Saint...

    2012

  13. [21]

    Time-changed spectrally positive L´ evy processes started from infinity

    Cl´ ement Foucart, Pei-Sen Li, and Xiaowen Zhou. Time-changed spectrally positive L´ evy processes started from infinity. Bernoulli, 27(2):1291–1318, 2021

    2021

  14. [22]

    Instantaneous support propagation for Λ-Fleming-Viot processes

    Thomas Hughes and Xiaowen Zhou. Instantaneous support propagation for Λ-Fleming-Viot processes. Stochastic Process. Appl., 155:535–560, 2023

    2023

  15. [23]

    The reversibility and an SPDE for the generalized Fleming-Viot processes with mutation

    Zenghu Li, Huili Liu, Jie Xiong, and Xiaowen Zhou. The reversibility and an SPDE for the generalized Fleming-Viot processes with mutation. Stochastic Process. Appl., 123(12):4129–4155, 2013

    2013

  16. [24]

    On the speed of coming down from infinity for Ξ-coalescent processes

    Vlada Limic. On the speed of coming down from infinity for Ξ-coalescent processes. Electron. J. Probab., 15:no. 8, 217–240, 2010

    2010

  17. [25]

    Second-order asymptotics for the block counting process in a class of regularly varying Λ-coalescents

    Vlada Limic and Anna Talarczyk. Second-order asymptotics for the block counting process in a class of regularly varying Λ-coalescents. Ann. Probab., 43(3):1419–1455, 2015

    2015

  18. [26]

    The compact support property for the Λ-Fleming-Viot process with underlying Brownian motion

    Huili Liu and Xiaowen Zhou. The compact support property for the Λ-Fleming-Viot process with underlying Brownian motion. Electron. J. Probab., 17:no. 73, 20, 2012

    2012

  19. [27]

    Some support properties for a class of Λ-Fleming-Viot processes

    Huili Liu and Xiaowen Zhou. Some support properties for a class of Λ-Fleming-Viot processes. Ann. Inst. Henri Poincar´ e Probab. Stat., 51(3):1076–1101, 2015

    2015

  20. [28]

    Exact modulus of continuities for Λ-Fleming-Viot processes with Brownian spatial motion

    Huili Liu and Xiaowen Zhou. Exact modulus of continuities for Λ-Fleming-Viot processes with Brownian spatial motion. J. Theor. Probab., 37(2):1710–1744, 2024

    2024

  21. [29]

    Coalescents with multiple collisions

    Jim Pitman. Coalescents with multiple collisions. Ann. Probab., 27(4):1870–1902, 1999

    1902

  22. [30]

    The general coalescent with asynchronous mergers of ancestral lines

    Serik Sagitov. The general coalescent with asynchronous mergers of ancestral lines. J. Appl. Probab. , 36(4):1116–1125, 1999

    1999

  23. [31]

    A necessary and sufficient condition for the Λ-coalescent to come down from infinity

    Jason Schweinsberg. A necessary and sufficient condition for the Λ-coalescent to come down from infinity. Electron. Comm. Probab., 5:1–11, 2000. Huili Liu: School of Mathematical Sciences, Hebei Normal University, Shijiazhuang, Hebei, 050024, China Email address : liuhuili@heb...

    2000

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.