Pith. sign in

REVIEW 3 major objections 4 minor 6 cited by

This paper argues that a neglected statistical prior—flat in distance modulus instead of flat in volume—biases galaxy distances low and the Hubble constant high, by up to ~8 km/s/Mpc in the CosmicFlows-4 sample.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

Implicit flat priors on distance moduli bias inferred distances low and the Hubble constant high — an ~8 km/s/Mpc (55σ) effect in CosmicFlows-4, and a ~1.6σ effect in SH0ES that is probably already absorbed by its bias corrections.

T0 review reviewed 2026-08-04 challenge →

load-bearing objection The statistical core is sound and genuinely useful; the CF4 headline rests on an unverifiable reading of other groups' catalogues, and the abstract oversells the SH0ES conclusion. the 3 major comments →

arxiv 2511.03394 v2 pith:C2FOSA35 submitted 2025-11-05 astro-ph.CO astro-ph.GA

The subtle statistics of the distance ladder: On the distance prior and selection effects

Harry Desmond , Richard Stiskalek , Jos\'e Antonio N\'ajera , Indranil Banik This is my paper
classification astro-ph.CO astro-ph.GA
keywords distance ladderHubble constantHubble tensiondistance priorselection effectsCosmicFlows-4SH0ESBayesian inference
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

The reading

The paper tries to establish that statistical choices in the distance ladder, not just astrophysics, can shift the inferred Hubble constant by amounts relevant to the Hubble tension. Distances are inferred, not measured, so they require priors; the physically motivated prior for volume-distributed galaxies is π(r)∝r², while many pipelines implicitly assume a flat prior on distance modulus, equivalent to π(r)∝1/r. The paper derives a simple formula for the resulting bias and shows it grows with √N, so large samples make the effect significant even when each distance is well measured. Applying the volume prior to CosmicFlows-4 shifts H0 down by ~8.3 km/s (a 55σ statistical shift), while selection effects—especially redshift selection—can partially or fully cancel this, and the effect on SH0ES is modest and likely already absorbed by simulation-based corrections. The paper's practical conclusion is that unbiased distance-ladder inference requires explicit modelling of both the volume prior and the selection function.

Core claim

The central claim, stated on the paper's own terms: because galaxies are uniformly distributed in three-dimensional space, the correct distance prior is π(r)∝r²; most distance-ladder pipelines implicitly use a flat prior on the distance modulus µ, equivalently π(r)∝1/r, which biases distances low and hence H0 high. In the toy model (standard candles of known absolute magnitude, Gaussian magnitude and redshift errors, power-law prior π(r)∝r^k on a known distance range), the MAP distance modulus shifts by (k+1)σm²/α relative to a pure likelihood fit, with α=5/ln10; moving from k=-1 (flat-µ) to k=2 (volume) changes ln H0 by -3σm²/α², i.e. -0.64% for σm=0.1 mag, and the bias relative to the post

What carries the argument

The engine of the paper is the one-dimensional negative log-posterior for the distance modulus, -lnP = Σ[(µ_i - m_i + M)²/(2σm²) - (k+1)µ_i/α], which turns a hierarchical distance-ladder inference into a two-term formula: the first term is the usual magnitude likelihood, the second is the sum of the volume prior's r^k and the r→µ Jacobian. From this the MAP shift and the H0 shift follow in closed form (Eqs. 8-13). The companion identity is the selection factor p(S=1|H0) ∝ H0^{-(1+k)} for redshift-selected samples, derived by marginalising over unobserved data; together these two expressions determine when a pipeline is biased and by how much, and they are validated with mocks.

Load-bearing premise

The paper's real-sample numbers rest on the assumption that CF4's published distance moduli are pure maximum-likelihood estimates with accurate uncertainties and that the published corrections do not already include the volume prior; if that assumption fails, the 8.3 km/s shift shrinks.

What would settle it

The paper's own mock test is the cleanest check: generate volume-limited samples with known H0 = 70, fit with both k=-1 and k=2 priors, and ask whether the k=-1 posterior is biased high at the predicted 0.35 km/s while k=2 is unbiased; the authors report this for 1500 mocks. For the real-data claim, the decisive observation is whether a reanalysis of CF4's raw Tully-Fisher measurements with an explicit forward model—volume prior plus a measured selection function—returns H0 close to 67 or close to 75 km/s/Mpc; if the flat-µ result survives a principled selection model, the headline shift is an

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • Any distance-ladder analysis that maximises a likelihood over distance moduli without a prior (or with a flat-µ prior) is implicitly using π(r)∝1/r and will overestimate H0 for volume- and magnitude-selected samples; the overestimate in km/s/Mpc is independent of sample size.
  • The relative bias grows as √N, so next-generation surveys with thousands of distance indicators will make this a dominant systematic unless the prior and selection are modelled.
  • For a perfectly redshift-selected sample with negligible redshift uncertainty, the flat-µ prior becomes unbiased because the selection factor cancels the volume prior—so the nature of selection, not just the prior, decides the direction of bias.
  • When applied to real data, the volume prior lowers CF4's H0 by ~8.3 km/s (55σ) in the no-selection/magnitude-selection limit, and lowers SH0ES's H0 by ~1.7 km/s (1.6σ), with the latter plausibly already corrected by simulation-based bias estimates.
  • Future distance-ladder samples should be designed or documented with homogeneous, known selection criteria; otherwise selection remains an irreducible systematic.

Where Pith is reading between the lines

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

  • If the volume-prior argument generalises, peculiar-velocity and growth-rate measurements derived from the same distance catalogues will also shift, since MAP distances change by ~12% for CF4-like uncertainties; this could bring independent distance and velocity measurements into better agreement or reveal new tensions.
  • A direct test of the paper's central identity is to infer the prior exponent k as a free hyperparameter from existing catalogues: mock data recover k≈2, so a real-data measurement of k significantly below 2 would indicate either a missing selection effect or a non-uniform source distribution.
  • The redshift-selection cancellation suggests a practical design principle: surveys that can enforce a strict, known redshift limit become insensitive to the volume prior, whereas magnitude-limited surveys must either adopt π(r)∝r² or simulate the selection.
  • The SH0ES pipeline's simulation-based corrections may be doing more work than previously appreciated; the paper's framework provides a way to check whether those corrections are complete by comparing against a forward model with the volume prior explicitly included.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 4 minor

Summary. The paper argues that distance-ladder analyses that implicitly use a uniform prior on distance moduli (equivalent to π(r)∝1/r) infer distances that are biased low and hence Hubble constants that are biased high when sources are uniformly distributed in volume. The authors derive analytic shifts for power-law distance priors (Eqs. 8–13), verify with 1500 Hamiltonian Monte Carlo mock datasets that π(r)∝r² is unbiased while π(r)∝1/r is biased, and extend the formalism to redshift, magnitude, and joint selection effects, finding that redshift selection can cancel the volume prior only in the zero-redshift-uncertainty limit. They apply the effect to CosmicFlows-4 and to SH0ES, reporting an 8.3 km/s/Mpc (55σ) shift for CF4 and a 1.6σ shift for SH0ES, while cautioning that both applications are illustrative and that selection modelling is essential.

Significance. The central methodological claim is classical but the paper gives a clean, self-contained derivation and a well-designed mock verification, making it a useful reference for the distance-ladder community. The SH0ES estimate (≈1.6σ, partly offset by selection corrections) is a valuable sanity check that the volume-prior effect is not a hidden resolution of the Hubble tension. However, the CF4 headline number is load-bearing for the paper's impact and rests on an external interpretation of how CF4's published distance moduli were constructed. If that interpretation is wrong, the 55σ shift is spurious; the paper's own caveats do not fully protect against this because the abstract still presents the number as a key finding.

major comments (3)
  1. [Sec. 4.1 and Abstract] The CF4 result (8.3 km/s/Mpc, 55σ) assumes that CF4's published distance moduli are pure maximum-likelihood estimates under a flat-µ prior (k=−1) and that the fn weighting in Springob et al. (2014)/Howlett et al. (2022) is only a selection correction. However, Tully et al. (2023, sec. 4.2) state that larger proposed distances are up-weighted to account for increased cosmological volume. At face value this contradicts the k=−1 baseline. The paper needs either to demonstrate from the CF4 construction that no volume weighting was applied, or to infer k from the CF4 data themselves; otherwise the headline shift should be explicitly downgraded from a key result to an assumption-dependent illustration.
  2. [Sec. 3.4 and Sec. 4.2.3] The phenomenological Lavaux model is used to argue that redshift selection is the more likely scenario for CF4, yielding H0≈78 km/s/Mpc and 'suggesting' a higher H0 is more realistic. But Fig. 4 shows this model is itself biased for redshift- and magnitude-selected samples, and the model parameters (p, R, q) are not directly tied to the actual selection function. The inferred H0≈78 can therefore not be used to diagnose the selection type. This weakens the conclusion in Sec. 6 that 'the truth is likely an amalgam' that lies between 67 and 75 km/s/Mpc; the current support for the higher end is largely from an 'indicative only' model.
  3. [Sec. 4.1 and Eq. (11)] The application of Eq. (11) to CF4 uses per-object σ_μ quoted in the catalogue and treats the distance measurements as uncorrelated. The paper correctly notes that systematic uncertainties and covariance in the scaling-relation parameters dominate the error budget (≈3 km/s/Mpc) and calls the estimate illustrative. Nevertheless, the abstract's '55σ' phrasing invites a statistical interpretation that is misleading, because the significance is computed against the statistical width only. The manuscript should either place the systematic caveat in the abstract or remove the 55σ claim from the abstract.
minor comments (4)
  1. [Data Availability] The code is said to be 'publicly available on GitHub/github-square', but this is not a usable URL or repository identifier. Please provide a full URL or DOI; otherwise the reproducibility claim cannot be verified.
  2. [Sec. 2.1 / Eq. (7)] The derivation of Eq. (7) assumes the δ-function constraint ri = c zi / H0, but the posterior in Eq. (6) also contains the bounds rmin ≤ ri ≤ rmax. In Sec. 3.1 the same setup is used without imposing bounds on the latent distances. Clarify which regime each equation refers to, since the two treatments lead to different conclusions about unbiasedness.
  3. [Sec. 5] The emulation of the volume prior for SH0ES SN distances through Eq. (38) assumes linear Hubble expansion and small σz, as acknowledged. It would be helpful to state how sensitive the 1.6σ shift is to relaxing these approximations, since the SH0ES sample extends to z where the cosmographic expansion terms are not negligible.
  4. [Sec. 4.1] Typographical/consistency: the abstract reports '8.3 km/s/Mpc (55σ)' while the text reports '≈55σ' and '8 km/s/Mpc' from Eq. (10). The precise value should be consistent across abstract, Sec. 4.1, and Sec. 6.

Circularity Check

0 steps flagged

No significant circularity: the core derivation is self-contained and benchmarked against external classical results.

full rationale

I walked the claimed derivation chain: the power-law distance prior (Eq. 4), the MAP calculation (Eqs. 6–13), the mock validation (Eq. 14, Fig. 2), the selection-effect integrals (Eqs. 18–23), and the CF4/SH0ES applications. The analytic shift Δln Ĥ0 = −σm²Δk/α² follows by direct differentiation of the explicit negative-log-posterior with no hidden fitted parameter, and it is independently re-derived in two limits in App. A. The mock test draws distances from a volume-uniform distribution and shows that k=2 is unbiased; this is a calibration check under the stated generative model, not a relabeling of input as output. The selection formalism in Eq. 18 is attributed to Kelly et al. (2008) and Stiskalek et al. (2025b), but the key redshift-selection cancellation (Eqs. 19–21) is derived in this paper, so the self-citation is not load-bearing. The CF4 headline shift is explicitly labelled 'illustrative only' (Sec. 4.1) and depends on interpreting external catalogues' fn weighting; that is an external-assumption/robustness concern, not a circular step. The SH0ES SN correction is constructed via Eq. 35 to emulate the analytic k-shift and is described as 'merely indicating the magnitude' (Sec. 5); it is an explicit sensitivity calculation rather than an independent prediction. The Lavaux (2016) prior is cited as an imperfect phenomenological model and is not used to ground the central conclusion. No uniqueness theorem is imported from the authors' prior work. Overall, the central results are self-contained and checked against Malmquist (1922) and Strauss & Willick (1995), so no circular step can be exhibited.

Axiom & Free-Parameter Ledger

3 free parameters · 5 axioms · 0 invented entities

The paper introduces no new physical entities. Its quantitative claims rest on: the domain assumption of a volume-uniform source distribution, the external Kelly et al. (2008) selection factor, and readings of third-party catalogues (CF4 moduli as k=−1 MLEs; SH0ES as χ² + bias corrections). Fitted inputs are limited to the assumed prior exponent k and the Lavaux (p, R, q) selection-model parameters, both transparently stated.

free parameters (3)
  • distance-prior exponent k = k = 2 (assumed); inferred ≈2 ± 0.07 in mock
    Headline results assume a power-law distance prior π(r)∝r^k (Eq. 4) with k=2 (volume) or k=−1 (flat-µ); a mock hyperparameter test infers k≈2, but real-data analyses fix k by assumption.
  • Lavaux phenomenological selection parameters (p, R, q) = p≈1.37, R≈56.6 Mpc, q≈1.34 (CF4 TFR fit)
    Fitted to CF4 data in Sec. 4.2.3/Fig. 5 to infer the effective selection; used to argue that redshift selection (which undoes the volume-prior shift) is the more likely CF4 scenario.
  • cosmographic parameters (q0, j0) = q0 = −0.595, j0 = 1
    Fixed following Tully et al. (2023) for the CF4 H₀ refit (Sec. 4.1); these inputs enter the predicted distance moduli (Eq. 29) but are not fitted here.
axioms (5)
  • domain assumption Galaxies are intrinsically uniformly distributed in volume, so the distance prior is π(r)∝r².
    Load-bearing prior for designating k=2 as 'unbiased'; invoked in Sec. 2 intro and baked into the mock truth (Eq. 14). The paper's central quantitative claims inherit this assumption.
  • standard math Selection effects modify the posterior by the factor [p(S=1|Λ)]^(−N).
    The selection formalism is imported from Kelly et al. (2008) via Eq. 18 and Stiskalek et al. (2025b); the redshift/magnitude selection results (Secs. 3.1–3.3) hold only to the extent this factor is the correct selection correction.
  • domain assumption CF4 catalogue distance moduli are uncorrected k=−1 MLEs with Gaussian, independent, correctly quoted uncertainties σµ.
    Sec. 4.1 applies Eq. 11 using published σµ (median 0.41) and treats the published moduli as if they carried no volume weighting; the paper argues this from catalogue documentation but cannot fully verify it.
  • domain assumption SH0ES uses a χ² estimator with σz propagation, and its simulation-based bias corrections plausibly already capture the volume prior.
    Sec. 5 states 'We assume that the basic inference method of SH0ES is the χ² estimator'; the conclusion that the volume effect is already absorbed is asserted with hedges ('should', 'plausibly'), not verified against the SH0ES pipeline.
  • ad hoc to paper The power-law prior with known r_min and r_max (Eq. 4) adequately idealizes real distance-ladder samples.
    The analytic framework requires known survey bounds; real samples (CF4, SH0ES) have neither clean volume limits nor documented homogeneous selection, so the quantitative claims inherit this idealization (acknowledged in Secs. 4 and 6).

reviewed 2026-08-04 · how reviews work

0 comments
Cite this review

Pith. "Pith review of The subtle statistics of the distance ladder: On the distance prior and selection effects." pith.science (2026). https://pith.science/paper/C2FOSA35

@misc{pith2026251103394,
  author       = {Pith},
  title        = {Pith review of: The subtle statistics of the distance ladder: On the distance prior and selection effects},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/C2FOSA35}},
  note         = {Machine review of arXiv:2511.03394}
}
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Statistical methodology is rarely considered significant in distance-ladder studies or a potential contributor to the Hubble tension. We suggest it should be, highlighting two appreciable issues. First, astronomical distances are inferred latent parameters, requiring a prior. We show that the (often implicit) uniform priors on distance moduli common to Bayesian distance-ladder analyses bias distances low due to objects being uniformly distributed in volume, which biases the Hubble constant high. Frequentist $\chi^2$ methods are unbiased for volume- or redshift-limited samples only if the redshift uncertainty (including peculiar velocities) vanishes, though simulation-based calibration can correct the bias. Second, in a Bayesian framework, selection effects introduce additional posterior factors describing the probability of objects entering the sample under the model. These partly counteract the volume prior, depending on the nature of the selection. After detailed analytic and mock-based studies, we quantify the volume-prior effect in the CosmicFlows-4 and SH0ES samples. Both use frequentist methods, so the effect appears as a potential estimator bias rather than a missing prior. The implied Hubble constant shifts are significant but must not be applied na\"{\i}vely -- principled selection modelling is also required, as we investigate explicitly for CosmicFlows-4. Both effects should already be captured by the SH0ES pipeline's simulation-based bias corrections. Our work highlights the crucial need to model both distances and selection accurately, either directly in a Bayesian forward model, or via post-hoc simulation-based corrections with realistic source and selection distributions. Such modelling requires samples with known, homogeneous selection criteria, which future surveys should prioritise.

Figures

Figures reproduced from arXiv: 2511.03394 by Harry Desmond, Indranil Banik, Jos\'e Antonio N\'ajera, Richard Stiskalek.

Figure 1
Figure 1. Figure 1: Directed acyclic graph depicting the simple distance ladder inference we use to illustrate the effect of the distance prior. be considered to be in units of 10 pc. The directed acyclic graph of this setup is shown in [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: The relative bias in H0 (in units of σ; Eq. 15) produced by the uniform-in-volume and uniform-in-µ distance priors across 1500 mock datasets. The agreement of the volume prior with the standard normal distribution shows that it is unbiased. predicts ∆Hˆ0 = 0.45 km/s/Mpc, but the average across the mock datasets is 0.35). Moreover, σ(H0) is increased by the redshift uncertainties, lowering the relative bias… view at source ↗
Figure 3
Figure 3. Figure 3: The average relative bias (solid lines) and 16th-84th percentile rangle (shaded bands) produced by the uniform-in-µ prior for a volume-limited sample as a function of σm, σz, and Ngal for fixed fiducial values of the other parameters, as indicated on each panel. further below, which simply maximises the likelihood of the observed magnitudes. Note that this equation assumes no bounds on the rˆi, violating t… view at source ↗
Figure 4
Figure 4. Figure 4: Relative bias B(H0) in the inferred Hubble constant of the Bayesian forward model and χ 2 estimator for the fiducial distance￾ladder setup, shown here as a function of σz separately for volume-, redshift-, and magnitude-limited samples. The lines show the median values over 200 mock datasets at each σz, while shaded bands show the 1σ range. We show results for the uniform-in-volume and uniform-in-µ distanc… view at source ↗
Figure 5
Figure 5. Figure 5: Corner plot (left) and distance histogram with overlaid best-fit prior (right) for the phenomenological selection model applied to the Tully–Fisher CF4 data without 4000 km s−1 cut. consequently, the inferred H0 would be between 67 and 75 km s−1 Mpc−1 . This would also be the case if joint magni￾tude and redshift selection were in operation. We leave quan￾tification of this for a future, more thorough reca… view at source ↗

discussion (0)

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

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Disentangling modified gravity and galaxy bias with field-level inference

    astro-ph.CO 2026-07 conditional novelty 6.0

    With fixed known initial phases, voxel-by-voxel Poisson likelihood on the galaxy number-counts field breaks the f(R)–bias degeneracy that power spectra cannot resolve, with voids and walls driving the gain.

  2. HALO II: Constraining Hubble constant $H_{0}$ through continuum delay fitting of Fairall 9

    astro-ph.GA 2026-06 unverdicted novelty 6.0

    Using simultaneous modeling of continuum lag-spectrum and broadband SED of Fairall 9 with the H0RIZON-AGN model, the authors obtain H0 = 72.4_{-3.7}^{+3.4} km s^{-1} Mpc^{-1}.

  3. Constraints on the gravitational potential from DESI DR2 BAO and its implications for the local void scenario

    astro-ph.CO 2026-06 conditional novelty 6.0

    A fixed gravitational-redshift contribution from a local void fits high-redshift data almost as well as ΛCDM, but the data currently favour z0=0.

  4. Constraints on the gravitational potential from DESI DR2 BAO and its implications for the local void scenario

    astro-ph.CO 2026-06 unverdicted novelty 5.0

    High-redshift datasets constrain the local void gravitational redshift parameter z0 to be consistent with zero but allow the value needed for Hubble tension solution.

  5. Hawai`i Supernova Flows: Bulk Flow Measurements using SNe Ia in the Optical and NIR

    astro-ph.CO 2026-06 unverdicted novelty 3.0

    Bulk flow measurements from Hawai`i Supernova Flows SNe Ia yield speeds of 100-400 km/s consistent with ΛCDM expectations at z ≲ 0.1.

  6. A Review on Resolving the Hubble Tension via Late-Universe Physics

    astro-ph.CO 2026-06 unverdicted novelty 2.0

    A review of late-universe models concludes that DESI BAO plus uncalibrated supernovae data indicate the Hubble tension originates in new low-redshift physics.

Reference graph

Works this paper leans on

59 extracted references · 5 canonical work pages · cited by 5 Pith papers · 1 internal anchor

  1. [1]

    Anton T., Clifton T., 2024, @doi [ ] 10.1088/1475-7516/2024/05/120 , https://ui.adsabs.harvard.edu/abs/2024JCAP...05..120A 2024, 120

    work page doi:10.1088/1475-7516/2024/05/120 2024

  2. [2]

    On the Impact of Cepheid Outliers on the Distance Ladder

    Becker M. R., Desmond H., Rozo E., Marshall P., Rykoff E. S., 2015, @doi [arXiv e-prints] 10.48550/arXiv.1507.07523 , https://ui.adsabs.harvard.edu/abs/2015arXiv150707523B p. arXiv:1507.07523

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1507.07523 2015

  3. [3]

    C., et al., 2019, @doi [ ] 10.1088/1538-3873/aaecbe , https://ui.adsabs.harvard.edu/abs/2019PASP..131a8002B 131, 018002

    Bellm E. C., et al., 2019, @doi [ ] 10.1088/1538-3873/aaecbe , https://ui.adsabs.harvard.edu/abs/2019PASP..131a8002B 131, 018002

    doi:10.1088/1538-3873/aaecbe 2019

  4. [4]

    Bingham E., et al., 2019, Journal of Machine Learning Research, 20, 1

    2019

  5. [5]

    P., Jensen J

    Blakeslee J. P., Jensen J. B., Ma C.-P., Milne P. A., Greene J. E., 2021, @doi [ ] 10.3847/1538-4357/abe86a , https://ui.adsabs.harvard.edu/abs/2021ApJ...911...65B 911, 65

    doi:10.3847/1538-4357/abe86a 2021

  6. [6]

    Breuval L., et al., 2024, @doi [ ] 10.3847/1538-4357/ad630e , https://ui.adsabs.harvard.edu/abs/2024ApJ...973...30B 973, 30

    doi:10.3847/1538-4357/ad630e 2024

  7. [7]

    R., et al., 2018, @doi [ ] 10.3847/1538-4357/aae51c , https://ui.adsabs.harvard.edu/abs/2018ApJ...869...56B 869, 56

    Burns C. R., et al., 2018, @doi [ ] 10.3847/1538-4357/aae51c , https://ui.adsabs.harvard.edu/abs/2018ApJ...869...56B 869, 56

    doi:10.3847/1538-4357/aae51c 2018

  8. [8]

    Calabrese E., et al., 2025, preprint, https://ui.adsabs.harvard.edu/abs/2025arXiv250314454C Arxiv ( @eprint arXiv 2503.14454 )

    Pith/arXiv arXiv 2025

  9. [9]

    Camphuis E., et al., 2025, preprint, https://ui.adsabs.harvard.edu/abs/2025arXiv250620707C Arxiv ( @eprint arXiv 2506.20707 )

    Pith/arXiv arXiv 2025

  10. [10]

    J., Lavaux G., Hudson M

    Carrick J., Turnbull S. J., Lavaux G., Hudson M. J., 2015, @doi [ ] 10.1093/mnras/stv547 , https://ui.adsabs.harvard.edu/abs/2015MNRAS.450..317C 450, 317

    doi:10.1093/mnras/stv547 2015

  11. [11]

    G., Miranda V., 2020, @doi [ ] 10.3847/1538-4357/ab7fb0 , https://ui.adsabs.harvard.edu/abs/2020ApJ...894...54D 894, 54

    Dhawan S., Brout D., Scolnic D., Goobar A., Riess A. G., Miranda V., 2020, @doi [ ] 10.3847/1538-4357/ab7fb0 , https://ui.adsabs.harvard.edu/abs/2020ApJ...894...54D 894, 54

    work page doi:10.3847/1538-4357/ab7fb0 2020

  12. [12]

    Di Valentino E., et al., 2021, @doi [Classical and Quantum Gravity] 10.1088/1361-6382/ac086d , https://ui.adsabs.harvard.edu/abs/2021CQGra..38o3001D 38, 153001

    doi:10.1088/1361-6382/ac086d 2021

  13. [13]

    Di Valentino E., et al., 2025, @doi [Physics of the Dark Universe] 10.1016/j.dark.2025.101965 , https://ui.adsabs.harvard.edu/abs/2025PDU....4901965D 49, 101965

    arXiv 2025

  14. [14]

    I., Steinmetz M., 2025, preprint, https://ui.adsabs.harvard.edu/abs/2025arXiv250722236D Arxiv ( @eprint arXiv 2507.22236 )

    Duangchan C., Valade A., Libeskind N. I., Steinmetz M., 2025, preprint, https://ui.adsabs.harvard.edu/abs/2025arXiv250722236D Arxiv ( @eprint arXiv 2507.22236 )

    arXiv 2025

  15. [15]

    M., 2023, @doi [ ] 10.1051/0004-6361/202346802 , https://ui.adsabs.harvard.edu/abs/2023A&A...678A.176D 678, A176

    Dupuy A., Courtois H. M., 2023, @doi [ ] 10.1051/0004-6361/202346802 , https://ui.adsabs.harvard.edu/abs/2023A&A...678A.176D 678, A176

    doi:10.1051/0004-6361/202346802 2023

  16. [16]

    M., Mortlock D

    Feeney S. M., Mortlock D. J., Dalmasso N., 2018, @doi [ ] 10.1093/mnras/sty418 , https://ui.adsabs.harvard.edu/abs/2018MNRAS.476.3861F 476, 3861

    doi:10.1093/mnras/sty418 2018

  17. [17]

    L., 2021, @doi [ ] 10.3847/1538-4357/ac0e95 , https://ui.adsabs.harvard.edu/abs/2021ApJ...919...16F 919, 16

    Freedman W. L., 2021, @doi [ ] 10.3847/1538-4357/ac0e95 , https://ui.adsabs.harvard.edu/abs/2021ApJ...919...16F 919, 16

    doi:10.3847/1538-4357/ac0e95 2021

  18. [18]

    L., Madore B

    Freedman W. L., Madore B. F., Hoyt T. J., Jang I. S., Lee A. J., Owens K. A., 2025, @doi [ ] 10.3847/1538-4357/adce78 , https://ui.adsabs.harvard.edu/abs/2025ApJ...985..203F 985, 203

    doi:10.3847/1538-4357/adce78 2025

  19. [19]

    B., 1992, @doi [Statistical Science] 10.1214/ss/1177011136 , 7, 457

    Gelman A., Rubin D. B., 1992, @doi [Statistical Science] 10.1214/ss/1177011136 , 7, 457

    arXiv 1992

  20. [20]

    S., Dhawan S., Uzsoy A

    Grayling M., Thorp S., Mandel K. S., Dhawan S., Uzsoy A. S. M., Boyd B. M., Hayes E. E., Ward S. M., 2024, @doi [ ] 10.1093/mnras/stae1202 , https://ui.adsabs.harvard.edu/abs/2024MNRAS.531..953G 531, 953

    doi:10.1093/mnras/stae1202 2024

  21. [21]

    H0DN Collaboration 2025, preprint, https://ui.adsabs.harvard.edu/abs/2025arXiv251023823H Arxiv ( @eprint arXiv 2510.23823 )

    Pith/arXiv arXiv 2025

  22. [22]

    R., et al., 2019, @doi [ ] 10.3847/1538-4357/ab13a3 , https://ui.adsabs.harvard.edu/abs/2019ApJ...876...15H 876, 15

    Hinton S. R., et al., 2019, @doi [ ] 10.3847/1538-4357/ab13a3 , https://ui.adsabs.harvard.edu/abs/2019ApJ...876...15H 876, 15

    doi:10.3847/1538-4357/ab13a3 2019

  23. [23]

    D., Gelman A., 2011, @doi [arXiv e-prints] 10.48550/arXiv.1111.4246 , https://ui.adsabs.harvard.edu/abs/2011arXiv1111.4246H p

    Hoffman M. D., Gelman A., 2011, @doi [arXiv e-prints] 10.48550/arXiv.1111.4246 , https://ui.adsabs.harvard.edu/abs/2011arXiv1111.4246H p. arXiv:1111.4246

    Pith/arXiv arXiv doi:10.48550/arxiv.1111.4246 2011

  24. [24]

    R., Colless M., Qin F., Lai Y., Tully R

    Howlett C., Said K., Lucey J. R., Colless M., Qin F., Lai Y., Tully R. B., Davis T. M., 2022, @doi [ ] 10.1093/mnras/stac1681 , https://ui.adsabs.harvard.edu/abs/2022MNRAS.515..953H 515, 953

    doi:10.1093/mnras/stac1681 2022

  25. [25]

    Hu J.-P., Wang F.-Y., 2023, @doi [Universe] 10.3390/universe9020094 , 9

    doi:10.3390/universe9020094 2023

  26. [26]

    arXiv:2508.15899

    Karchev K., Trotta R., Jimenez R., 2025, arXiv e-prints, https://ui.adsabs.harvard.edu/abs/2025arXiv250815899K p. arXiv:2508.15899

    Pith/arXiv arXiv 2025

  27. [27]

    C., Fan X., Vestergaard M., 2008, @doi [ ] 10.1086/589501 , https://ui.adsabs.harvard.edu/abs/2008ApJ...682..874K 682, 874

    Kelly B. C., Fan X., Vestergaard M., 2008, @doi [ ] 10.1086/589501 , https://ui.adsabs.harvard.edu/abs/2008ApJ...682..874K 682, 874

    doi:10.1086/589501 2008

  28. [28]

    Kessler R., Scolnic D., 2017, @doi [ ] 10.3847/1538-4357/836/1/56 , https://ui.adsabs.harvard.edu/abs/2017ApJ...836...56K 836, 56

    doi:10.3847/1538-4357/836/1/56 2017

  29. [29]

    B., Anand G

    Kourkchi E., Tully R. B., Anand G. S., Courtois H. M., Dupuy A., Neill J. D., Rizzi L., Seibert M., 2020a, @doi [ ] 10.3847/1538-4357/ab901c , https://ui.adsabs.harvard.edu/abs/2020ApJ...896....3K 896, 3

    doi:10.3847/1538-4357/ab901c

  30. [30]

    Kourkchi E., et al., 2020b, @doi [ ] 10.3847/1538-4357/abb66b , https://ui.adsabs.harvard.edu/abs/2020ApJ...902..145K 902, 145

    doi:10.3847/1538-4357/abb66b

  31. [31]

    B., Courtois H

    Kourkchi E., Tully R. B., Courtois H. M., Dupuy A., Guinet D., 2022, @doi [ ] 10.1093/mnras/stac303 , https://ui.adsabs.harvard.edu/abs/2022MNRAS.511.6160K 511, 6160

    doi:10.1093/mnras/stac303 2022

  32. [32]

    Lavaux G., 2016, @doi [ ] 10.1093/mnras/stv2915 , https://ui.adsabs.harvard.edu/abs/2016MNRAS.457..172L 457, 172

    doi:10.1093/mnras/stv2915 2016

  33. [33]

    M., Burstein D., Davies R

    Lynden-Bell D., Faber S. M., Burstein D., Davies R. L., Dressler A., Terlevich R. J., Wegner G., 1988, @doi [ApJ] 10.1086/166066 , https://ui.adsabs.harvard.edu/abs/1988ApJ...326...19L 326, 19

    doi:10.1086/166066 1988

  34. [34]

    G., 1922, Meddelanden fran Lunds Astronomiska Observatorium Serie I, https://ui.adsabs.harvard.edu/abs/1922MeLuF.100....1M 100, 1

    Malmquist K. G., 1922, Meddelanden fran Lunds Astronomiska Observatorium Serie I, https://ui.adsabs.harvard.edu/abs/1922MeLuF.100....1M 100, 1

    1922

  35. [35]

    S., Wood-Vasey W

    Mandel K. S., Wood-Vasey W. M., Friedman A. S., Kirshner R. P., 2009, @doi [ ] 10.1088/0004-637X/704/1/629 , https://ui.adsabs.harvard.edu/abs/2009ApJ...704..629M 704, 629

    doi:10.1088/0004-637x/704/1/629 2009

  36. [36]

    S., Narayan G., Kirshner R

    Mandel K. S., Narayan G., Kirshner R. P., 2011, @doi [ ] 10.1088/0004-637X/731/2/120 , https://ui.adsabs.harvard.edu/abs/2011ApJ...731..120M 731, 120

    doi:10.1088/0004-637x/731/2/120 2011

  37. [37]

    S., Scolnic D

    Mandel K. S., Scolnic D. M., Shariff H., Foley R. J., Kirshner R. P., 2017, @doi [ ] 10.3847/1538-4357/aa6038 , https://ui.adsabs.harvard.edu/abs/2017ApJ...842...93M 842, 93

    doi:10.3847/1538-4357/aa6038 2017

  38. [38]

    S., Thorp S., Narayan G., Friedman A

    Mandel K. S., Thorp S., Narayan G., Friedman A. S., Avelino A., 2022, @doi [ ] 10.1093/mnras/stab3496 , https://ui.adsabs.harvard.edu/abs/2022MNRAS.510.3939M 510, 3939

    doi:10.1093/mnras/stab3496 2022

  39. [39]

    C., Trotta R., Berkes P., Starkman G

    March M. C., Trotta R., Berkes P., Starkman G. D., Vaudrevange P. M., 2011, @doi [ ] 10.1111/j.1365-2966.2011.19584.x , https://ui.adsabs.harvard.edu/abs/2011MNRAS.418.2308M 418, 2308

    arXiv 2011

  40. [40]

    S., Jenkins A., 2025, @doi [ ] 10.1093/mnras/staf767 , https://ui.adsabs.harvard.edu/abs/2025MNRAS.540..716M 540, 716

    McAlpine S., Jasche J., Ata M., Lavaux G., Stiskalek R., Frenk C. S., Jenkins A., 2025, @doi [ ] 10.1093/mnras/staf767 , https://ui.adsabs.harvard.edu/abs/2025MNRAS.540..716M 540, 716

    doi:10.1093/mnras/staf767 2025

  41. [41]

    A., Desmond H., 2025, @doi [ ] 10.1093/mnras/staf1005 , https://ui.adsabs.harvard.edu/abs/2025MNRAS.541..671N 541, 671

    N \'a jera J. A., Desmond H., 2025, @doi [ ] 10.1093/mnras/staf1005 , https://ui.adsabs.harvard.edu/abs/2025MNRAS.541..671N 541, 671

    work page doi:10.1093/mnras/staf1005 2025

  42. [42]

    Nimonkar H., Mukherjee S., 2024, @doi [ ] 10.1093/mnras/stad3256 , https://ui.adsabs.harvard.edu/abs/2024MNRAS.527.2152N 527, 2152

    work page doi:10.1093/mnras/stad3256 2024

  43. [43]

    arXiv:1912.11554

    Phan D., Pradhan N., Jankowiak M., 2019, @doi [arXiv e-prints] 10.48550/arXiv.1912.11554 , https://ui.adsabs.harvard.edu/abs/2019arXiv191211554P p. arXiv:1912.11554

    Pith/arXiv arXiv doi:10.48550/arxiv.1912.11554 2019

  44. [44]

    Planck Collaboration 2020, @doi [ ] 10.1051/0004-6361/201833910 , https://ui.adsabs.harvard.edu/abs/2020A&A...641A...6P 641, A6

    doi:10.1051/0004-6361/201833910 2020

  45. [45]

    Popovic B., Brout D., Kessler R., Scolnic D., Lu L., 2021, @doi [ ] 10.3847/1538-4357/abf14f , https://ui.adsabs.harvard.edu/abs/2021ApJ...913...49P 913, 49

    doi:10.3847/1538-4357/abf14f 2021

  46. [46]

    G., et al., 2022, @doi [ ] 10.3847/2041-8213/ac5c5b , https://ui.adsabs.harvard.edu/abs/2022ApJ...934L...7R 934, L7

    Riess A. G., et al., 2022, @doi [ ] 10.3847/2041-8213/ac5c5b , https://ui.adsabs.harvard.edu/abs/2022ApJ...934L...7R 934, L7

    doi:10.3847/2041-8213/ac5c5b 2022

  47. [47]

    Rigault M., et al., 2025, @doi [ ] 10.1051/0004-6361/202450388 , https://ui.adsabs.harvard.edu/abs/2025A&A...694A...1R 694, A1

    doi:10.1051/0004-6361/202450388 2025

  48. [48]

    Rubin D., et al., 2015, @doi [ ] 10.1088/0004-637X/813/2/137 , https://ui.adsabs.harvard.edu/abs/2015ApJ...813..137R 813, 137

    doi:10.1088/0004-637x/813/2/137 2015

  49. [49]

    Schombert J., McGaugh S., Lelli F., 2020, @doi [ ] 10.3847/1538-3881/ab9d88 , https://ui.adsabs.harvard.edu/abs/2020AJ....160...71S 160, 71

    doi:10.3847/1538-3881/ab9d88 2020

  50. [50]

    M., et al., 2014, @doi [ ] 10.1093/mnras/stu1743 , https://ui.adsabs.harvard.edu/abs/2014MNRAS.445.2677S 445, 2677

    Springob C. M., et al., 2014, @doi [ ] 10.1093/mnras/stu1743 , https://ui.adsabs.harvard.edu/abs/2014MNRAS.445.2677S 445, 2677

    doi:10.1093/mnras/stu1743 2014

  51. [51]

    J., Bartlett D

    Stiskalek R., Desmond H., Devriendt J., Slyz A., Lavaux G., Hudson M. J., Bartlett D. J., Courtois H. M., 2025a, @doi [arXiv e-prints] 10.48550/arXiv.2502.00121 , https://ui.adsabs.harvard.edu/abs/2025arXiv250200121S p. arXiv:2502.00121

    doi:10.48550/arxiv.2502.00121

  52. [52]

    arXiv:2509.09665

    Stiskalek R., Desmond H., Tsaprazi E., Heavens A., Lavaux G., McAlpine S., Jasche J., 2025b, arXiv e-prints, https://ui.adsabs.harvard.edu/abs/2025arXiv250909665S p. arXiv:2509.09665

    Pith/arXiv arXiv

  53. [53]

    Stiskalek R., Desmond H., Banik I., 2025c, @doi [ ] 10.1093/mnras/staf1571 , https://ui.adsabs.harvard.edu/abs/2025MNRAS.543.1556S 543, 1556

    doi:10.1093/mnras/staf1571

  54. [54]

    A., Willick J

    Strauss M. A., Willick J. A., 1995, @doi [ ] 10.1016/0370-1573(95)00013-7 , https://ui.adsabs.harvard.edu/abs/1995PhR...261..271S 261, 271

    doi:10.1016/0370-1573(95)00013-7 1995

  55. [55]

    Tristram M., et al., 2024, @doi [ ] 10.1051/0004-6361/202348015 , https://ui.adsabs.harvard.edu/abs/2024A&A...682A..37T 682, A37

    doi:10.1051/0004-6361/202348015 2024

  56. [56]

    B., et al., 2013, @doi [ ] 10.1088/0004-6256/146/4/86 , https://ui.adsabs.harvard.edu/abs/2013AJ....146...86T 146, 86

    Tully R. B., et al., 2013, @doi [ ] 10.1088/0004-6256/146/4/86 , https://ui.adsabs.harvard.edu/abs/2013AJ....146...86T 146, 86

    doi:10.1088/0004-6256/146/4/86 2013

  57. [57]

    B., et al., 2023, @doi [ ] 10.3847/1538-4357/ac94d8 , https://ui.adsabs.harvard.edu/abs/2023ApJ...944...94T 944, 94

    Tully R. B., et al., 2023, @doi [ ] 10.3847/1538-4357/ac94d8 , https://ui.adsabs.harvard.edu/abs/2023ApJ...944...94T 944, 94

    doi:10.3847/1538-4357/ac94d8 2023

  58. [58]

    Watkins R., et al., 2023, @doi [ ] 10.1093/mnras/stad1984 , https://ui.adsabs.harvard.edu/abs/2023MNRAS.524.1885W 524, 1885

    doi:10.1093/mnras/stad1984 2023

  59. [59]

    G., Stahl B

    de Jaeger T., Galbany L., Riess A. G., Stahl B. E., Shappee B. J., Filippenko A. V., Zheng W., 2022, @doi [ ] 10.1093/mnras/stac1661 , https://ui.adsabs.harvard.edu/abs/2022MNRAS.514.4620D 514, 4620

    doi:10.1093/mnras/stac1661 2022

This paper was first reviewed by deepseek-v4-flash on August 4, 2026.