Pith. sign in

REVIEW 3 major objections 5 minor 80 references

Cosmological black-hole binaries typically merge in about 1 Gyr, short enough to match pulsar-timing signals.

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 ↗

T0 review · grok-4.5

2026-07-10 17:34 UTC pith:A2IQOM53

load-bearing objection Solid, usable scaling relations from realistic cosmological ICs; the ~1 Gyr PTA-friendly peak is real for the gas-poor major-merger population they actually trained on, but rests on N=30 linear fits that should not be over-sold as universal. the 3 major comments →

arxiv 2607.07813 v1 pith:A2IQOM53 submitted 2026-07-08 astro-ph.GA

Scaling Relations for Binary Black Hole Merger Times from Cosmological Initial Conditions

Thibaut L. Francois , Alessia Gualandris , Walter Dehnen This is my paper
classification astro-ph.GA
keywords massive black hole binariesgravitational wave backgroundpulsar timing arraysdynamical frictionstellar hardeningscaling relationsgalaxy mergersIllustrisTNG
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.

Pulsar timing arrays have detected a nanohertz gravitational-wave background whose amplitude sits a little high compared with standard models of massive black-hole binaries. The missing ingredient is how long those binaries take to merge once their host galaxies collide. This paper takes 30 major galaxy mergers straight from a cosmological simulation, re-simulates them at high resolution, and follows the black holes all the way to coalescence. The encounters and the resulting binaries are almost always highly eccentric. From the runs the authors extract simple scaling relations that link dynamical-friction, hardening and total merger times to host-galaxy and orbital properties. When the relations are applied back to the full cosmological merger catalogue, galaxy mergers finish in roughly 0.7 Gyr and black holes coalesce in roughly 1 Gyr. Those short delays mean binaries evolve efficiently enough to produce the observed gravitational-wave background.

Core claim

Applying the scaling relations derived from high-resolution re-simulations of 30 IllustrisTNG major mergers yields a galaxy-merger time distribution that peaks near 0.67 Gyr and a total black-hole coalescence time that peaks near 1.0 Gyr, implying efficient binary evolution consistent with current pulsar-timing-array constraints.

What carries the argument

Empirical power-law scaling relations (Table 3) that express dynamical-friction, hardening and total coalescence times as linear combinations of a few host-galaxy and orbital parameters (chiefly the encounter semi-major axis, eccentricity and central densities or masses).

Load-bearing premise

The 30 carefully selected gas-poor major mergers are representative enough that the fitted linear relations can be extrapolated to the whole cosmological merger population.

What would settle it

A larger suite of high-resolution re-simulations that includes gas-rich or minor mergers and finds systematically longer coalescence times (several Gyr or more) would falsify the claim that the typical delay is only ~1 Gyr.

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

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 / 5 minor

Summary. The paper extracts 30 major, gas-poor galaxy mergers from IllustrisTNG (z≤1, q★>1/4, chirp mass >10^8 M⊙), re-samples them at high resolution with Agama, and evolves them with the Griffin N-body code through dynamical friction and stellar hardening. A semi-analytic Peters integration with MCMC uncertainty on the hardening rate then carries each binary to coalescence. Cosmological encounters and the resulting binaries are found to be highly eccentric (e0 peak ~0.94, eb peak ~0.91, ePTA peak ~0.992). Empirical log-linear scaling relations (Table 3) are derived for dynamical-friction, hardening and total coalescence times via exhaustive subset selection, VIF multicollinearity cuts and LOOCV. Applying the most practical relations (c and q) to a larger TNG sample of 174 mergers yields galaxy-merger times peaking at ~0.67 Gyr and total BH coalescence times peaking at ~1.0 Gyr, presented as consistent with PTA constraints on efficient binary evolution.

Significance. If the short coalescence-time distribution is robust, the work supplies a concrete, observationally motivated subgrid prescription that can replace the common “prompt merger” or purely analytic hardening recipes used in cosmological GWB forecasts. The use of genuine cosmological initial conditions (highly eccentric, DM-stripped secondaries) rather than idealised equal-halo setups is a clear advance over Holley-Bockelmann et al. (2025). Strengths include the careful mass-refinement and mass-dependent softening scheme, the MCMC treatment of s(t), the VIF/LOOCV selection protocol, and the explicit practical guide in §5.4. The eccentricity statistics at PTA entry are also of direct interest for waveform modelling.

major comments (3)
  1. The central claim of §4.5 and Fig. 13 (peaks at ~0.67 Gyr and ~1.0 Gyr) rests on extrapolating relations (c) and (q) fitted exclusively to the N=30 gas-poor training set. Relation (q) has R²=0.86 and LOOCV RMSE=0.37 dex; the pure-global hardening relation (j) drops to R²=0.45. The paper acknowledges the gas cut and sample size in §5.3 but does not quantify how the peak location shifts under leave-k-out subsets, bootstrap resampling of the 30 runs, or inclusion of systems that fail the 30 % gas cut. A short sensitivity test is needed before the PTA-consistency statement can be regarded as secure.
  2. §4.4.4 and Table 3: once binary-orbit parameters are unavailable, the hardening timescale is only weakly constrained (relation j, R²=0.45, Δ=0.38). Because dynamical friction dominates the total time, the coalescence relations remain usable, but the residual stochasticity in eb (explicitly discussed in §5.2) is not folded into the Gaussian mixture of Fig. 13. Propagating that extra scatter (or at least quoting an enlarged uncertainty band) would strengthen the cosmological application.
  3. §2.1.2 and §5.3: the training sample is restricted to major stellar mergers with gas mass <30 % of stellar mass inside 2 rhm. The larger catalogue of 174 systems still excludes gas-rich mergers. Given that PTA sources may include a non-negligible gas-rich fraction, the claim that the ~1 Gyr peak is representative of the full PTA-relevant population needs either a quantitative bound or a clearer statement of the restricted domain of applicability.
minor comments (5)
  1. Fig. 12 panels are dense; adding a one-to-one residual histogram or a colour-coding by TNG volume would help the reader assess residual trends.
  2. Table 3: the units of each coefficient are not always obvious from the log10 expressions; a short column or footnote listing the physical units of every predictor would improve usability for subgrid implementers.
  3. §2.2.5: the switch from εBH,df=10 pc to εBH,bin=2 pc after the third pericentre is sensible, but a one-sentence justification that the choice does not affect the measured hardening rate would be welcome.
  4. Appendix B: the mild offset of the TNG black holes below the Reines & Volonteri relation is noted but not discussed; a brief remark on whether this biases the chirp-mass selection would be useful.
  5. A few typographical inconsistencies remain (e.g., “parametres” vs “parameters”, occasional missing spaces around ~).

Circularity Check

0 steps flagged

No load-bearing circularity: scaling relations are empirical fits to 30 independent high-resolution runs, then applied out-of-sample to a larger TNG catalogue; only a non-essential self-citation on resolution appears.

full rationale

The derivation chain is: (i) extract 30 gas-poor major mergers from TNG, re-simulate with Griffin to measure Δt_df, Δt_h and t_coal, (ii) perform best-subset linear regression in log-space on host/orbital parameters (Table 3, relations a–q) with VIF and LOOCV controls, (iii) apply the resulting formulae (chiefly c and q) to an independent larger sample of 174 TNG mergers whose parameters are measured from the same simulation suite but without the high-resolution re-simulations. The population peaks (~0.67 Gyr galaxy merger, ~1.0 Gyr BH coalescence) are therefore genuine out-of-sample predictions, not tautological re-statements of the fit. The only self-citation of overlapping authors (Gualandris et al. 2026) concerns numerical stochasticity of e_b and is used solely as a qualitative resolution guideline; it does not enter the scaling coefficients or the final distributions. No equation reduces a claimed prediction to a fitted input by construction, no uniqueness theorem is imported, and no ansatz is smuggled. Residual concerns about sample size, gas cut and extrapolation are scientific limitations, not circularity.

Axiom & Free-Parameter Ledger

4 free parameters · 4 axioms · 0 invented entities

The central claim rests on standard gravitational dynamics plus a set of empirical linear regressions whose coefficients are free parameters fitted to 30 simulations; no new physical entities are postulated.

free parameters (4)
  • coefficients of the 17 scaling relations (Table 3)
    Each relation is a multi-linear fit in log-space; the numerical prefactors and exponents are determined by least-squares on the 30-run sample and are not predicted a priori.
  • mass-refinement shell multipliers and population fractions (Table 2)
    Chosen by hand (1,3,10,40 and 0.05/0.45/0.35/0.15) to balance central resolution against outer-particle mass; different choices alter the particle spectrum.
  • softening lengths ε0=20 pc, εBH,df=10 pc, εBH,bin=2 pc
    Reference softenings set by numerical experiment; they control the collisionality of the binary and the outer halo.
  • gas-mass cut <30 % of stellar mass
    Ad-hoc selection threshold that defines the 30-run training set; changing it would alter the sample.
axioms (4)
  • domain assumption Peters (1964) orbit-averaged GW equations remain valid down to 10 Schwarzschild radii
    Used to terminate the semi-analytic integration (§3.4); verified only a posteriori for most runs.
  • ad hoc to paper Stellar hardening rate s(t) can be represented by a decaying exponential fitted by MCMC
    Functional form chosen for convenience; alternative binning or functional forms change the inferred coalescence time within the quoted 16–84 percentiles.
  • domain assumption Collisionless N-body dynamics with mass-dependent softening adequately captures dynamical friction and three-body hardening in gas-poor systems
    Standard assumption of the Griffin runs; gas and collisional stellar evolution are omitted by construction.
  • ad hoc to paper Linear regression in log-space after standardisation is sufficient to capture the dominant power-law dependencies
    Higher-order or non-separable terms are not explored because N=30 is too small (§5.3).

pith-pipeline@v1.1.0-grok45 · 38272 in / 2851 out tokens · 40512 ms · 2026-07-10T17:34:44.477799+00:00 · methodology

0 comments
read the original abstract

Recent evidence from Pulsar Timing Arrays (PTAs) for a nanohertz gravitational wave background is broadly consistent with theoretical expectations from a population of massive black hole binaries (MBHBs), although the inferred amplitude appears somewhat higher than predicted by standard models. Interpreting these observations requires a robust understanding of the merger timescales of MBHBs, and of their connection to host galaxy properties. In this work, we investigate the evolution of MBHBs selected from cosmological galaxy mergers in the IllustrisTNG simulation. We re-simulate these systems at high resolution using the N-body code Griffin to accurately resolve the dynamical friction and stellar hardening phases, and follow their evolution to coalescence with a semi-analytical model. We find that cosmological galaxy encounters and the resulting MBHBs are typically highly eccentric. We characterise the distribution of binary eccentricities at formation and at entry into the PTA band, and quantify the corresponding residence times. We identify the key parameters governing the duration of the different stages of MBHB evolution, and derive scaling relations linking galaxy and orbital properties to dynamical friction, hardening, and total coalescence times. These relations provide a framework for subgrid prescriptions in cosmological simulations. Applying these scaling relations to the full IllustrisTNG merger population, we infer the probability distributions of galaxy merger and black hole coalescence times. We find that galaxy mergers typically complete within $\sim 0.7$ Gyr, while the total black hole coalescence time is $\sim 1.0$ Gyr. These short timescales imply efficient binary evolution, consistent with current PTA constraints.

Figures

Figures reproduced from arXiv: 2607.07813 by Alessia Gualandris, Thibaut L. Francois, Walter Dehnen.

Figure 1
Figure 1. Figure 1: Density profile fits for galaxies extracted from the IllustrisTNG simulations. The original TNG density profiles are shown as thin, dark lines, while the corresponding best-fitting truncated spheroidal models are displayed as thick, lighter lines. Dark matter and stellar components are shown in blue and red, respectively. Each column presents an example pair of galaxies drawn from one of the three Illustri… view at source ↗
Figure 2
Figure 2. Figure 2: Probability density distributions of the masses of the selected IllustrisTNG galaxy mergers. Blue histograms show the dark matter (DM) masses of the first progenitor (FP), the second progenitor (SP), and their dark matter mass ratio. Red histograms show the corresponding stellar (★) masses and stellar mass ratio, while green histograms show the black hole (BH) masses of the central black holes in the FP an… view at source ↗
Figure 3
Figure 3. Figure 3: Probability density distributions of the orbital parameters of the selected IllustrisTNG encounters. The left panel shows one-dimensional probability density histograms of the orbital eccentricity, semi-major axis, pericentre, and apocentre, with the latter two normalised by the half-mass radius of the first progenitor (𝑟hm,FP). The peak values of the eccentricity and semi-major axis distributions are indi… view at source ↗
Figure 6
Figure 6. Figure 6: Evolution of the black hole binary in merger 7 as a function of time, computed with the semi-analytic model (SAM) described in Section 2.3. The top panel shows the semi-major axis, 𝑎, while the bottom panel shows the eccentricity, 𝑒. Black points correspond to the 𝑁-body simulation data, and colored lines to SAM integrations. Different colors indicate different initial times for the SAM integration, while … view at source ↗
Figure 5
Figure 5. Figure 5: Probability density distributions of binary eccentricities. The left panel shows the distribution of 𝑒b at binary formation, while the right panel shows the distribution of 𝑒h at the hard separation time. Solid lines indicate kernel density estimates, with their peak values marked, while dashed lines denote the median values. The distributions peak at 𝑒b = 0.91 and 𝑒h = 0.93, with corresponding median valu… view at source ↗
Figure 7
Figure 7. Figure 7: Timescale decomposition of the three phases of binary evolution for the 30 mergers. The red segment represents the duration of the dynamical friction phase, Δ𝑡df = 𝑡b, the orange segment the hardening phase, Δ𝑡h = 𝑡gw − 𝑡b, and the blue segment the gravitational wave driven phase, Δ𝑡gw = 𝑡coal − 𝑡gw. Black boxes indicate the uncertainty on the coalescence time, defined by the interval between the 16th and … view at source ↗
Figure 8
Figure 8. Figure 8: Probability density distributions of timescale ratios. The left panel shows the distribution ofΔ𝑡df/Δ𝑡h, while the right panel showsΔ𝑡h/Δ𝑡gw. Peak values of the kernel density estimates are indicated. These distributions show that the dynamical friction phase is typically the longest, with a characteristic duration about twice that of the hardening phase, which itself is on average about twice as long as t… view at source ↗
Figure 10
Figure 10. Figure 10: Probability density distributions of the time spent in the PTA band, Δ𝑡PTA (left), and the eccentricity at entry into the PTA band, 𝑒PTA (right). Solid lines show kernel density estimates, with peak values indicated. These distributions show that binaries typically enter the PTA band with very high eccentricities, peaking at 𝑒PTA = 0.992, and remain in the band for ∼ 131 Myr. The median residence time and… view at source ↗
Figure 9
Figure 9. Figure 9: Evolution of the binary from merger 7 within the PTA band. The top panel shows the peak gravitational-wave frequency, 𝑓p, as a function of time. The middle and bottom panels show the evolution of the semi-major axis, 𝑎, and eccentricity, 𝑒, respectively, as functions of 𝑓p. In all panels, the PTA band (10−9 − 10−7 Hz) is highlighted by the shaded blue region. Different colours correspond to different initi… view at source ↗
Figure 11
Figure 11. Figure 11: shows the correlation between the hardening rate (𝑠˜) and the central stellar density of the remnant (𝜌rem(𝑟inf)) 9 . This provides a useful consistency check, ensuring that the hardening process is physically well captured in our simulations. The grey diamond mark￾ers correspond to the data from Holley-Bockelmann et al. (2025), which overlap with the region of the (𝑠, 𝜌) parameter space spanned by our si… view at source ↗
Figure 12
Figure 12. Figure 12: Measured versus predicted timescales for the scaling relations listed in [PITH_FULL_IMAGE:figures/full_fig_p016_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: Distributions of predicted galaxy merger times Δ𝑡df (left) and total BH coalescence times 𝑡coal (right) for the 174 mergers selected from IllustrisTNG (TNG50, TNG100, and TNG300) at 𝑧 ≤ 1, estimated using relations (c) and (q). The distributions are constructed as an equal-weight Gaussian mixture, with one component centred on each individual prediction and a fixed width set by the intrinsic scatter of th… view at source ↗

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

80 extracted references · 80 canonical work pages · 2 internal anchors

  1. [1]

    , keywords =

    E pur si muove: Galilean-invariant cosmological hydrodynamical simulations on a moving mesh. , keywords =

    work page

  2. [2]

    Planck 2015 results. XIII. Cosmological parameters. , keywords =

    work page 2015

  3. [3]

    , keywords =

    Simulating galaxy formation with black hole driven thermal and kinetic feedback. , keywords =

    work page

  4. [4]

    , keywords =

    Supermassive black holes and their feedback effects in the IllustrisTNG simulation. , keywords =

    work page

  5. [5]

    , keywords =

    Simulating galaxy formation with the IllustrisTNG model. , keywords =

    work page

  6. [6]

    , keywords =

    Properties of galaxies reproduced by a hydrodynamic simulation. , keywords =

    work page

  7. [7]

    , keywords =

    Introducing the Illustris project: the evolution of galaxy populations across cosmic time. , keywords =

    work page

  8. [8]

    , keywords =

    The Illustris simulation: the evolving population of black holes across cosmic time. , keywords =

    work page

  9. [9]

    , keywords =

    First results from the IllustrisTNG simulations: matter and galaxy clustering. , keywords =

    work page

  10. [10]

    , keywords =

    First results from the IllustrisTNG simulations: the stellar mass content of groups and clusters of galaxies. , keywords =

    work page

  11. [11]

    , keywords =

    First results from the IllustrisTNG simulations: a tale of two elements - chemical evolution of magnesium and europium. , keywords =

    work page

  12. [12]

    , keywords =

    First results from the IllustrisTNG simulations: the galaxy colour bimodality. , keywords =

    work page

  13. [13]

    , keywords =

    First results from the IllustrisTNG simulations: radio haloes and magnetic fields. , keywords =

    work page

  14. [14]

    , keywords =

    First results from the TNG50 simulation: galactic outflows driven by supernovae and black hole feedback. , keywords =

    work page

  15. [15]

    , keywords =

    First results from the TNG50 simulation: the evolution of stellar and gaseous discs across cosmic time. , keywords =

    work page

  16. [16]

    Computational Astrophysics and Cosmology , keywords =

    The IllustrisTNG simulations: public data release. Computational Astrophysics and Cosmology , keywords =

    work page

  17. [17]

    , keywords =

    The merger rate of galaxies in the Illustris simulation: a comparison with observations and semi-empirical models. , keywords =

    work page

  18. [18]

    , keywords =

    Massive black hole evolution models confronting the n-Hz amplitude of the stochastic gravitational wave background. , keywords =

    work page

  19. [19]

    , keywords =

    The stochastic gravitational-wave background from massive black hole binary systems: implications for observations with Pulsar Timing Arrays. , keywords =

    work page

  20. [20]

    , keywords =

    Gravitational waves from resolvable massive black hole binary systems and observations with Pulsar Timing Arrays. , keywords =

    work page

  21. [21]

    , keywords =

    AGAMA: action-based galaxy modelling architecture. , keywords =

    work page

  22. [22]

    arXiv e-prints , keywords =

    Converging on the eccentricity of massive black hole binaries in galactic mergers. arXiv e-prints , keywords =

    work page

  23. [23]

    , keywords =

    Reviving stochasticity: uncertainty in SMBH binary eccentricity is unavoidable. , keywords =

    work page

  24. [24]

    , keywords =

    A Very Fast and Momentum-conserving Tree Code. , keywords =

    work page

  25. [25]

    Computational Astrophysics and Cosmology , keywords =

    A fast multipole method for stellar dynamics. Computational Astrophysics and Cosmology , keywords =

    work page

  26. [26]

    , keywords =

    Relations between Central Black Hole Mass and Total Galaxy Stellar Mass in the Local Universe. , keywords =

    work page

  27. [27]

    , keywords =

    Orbital parameters of merging dark matter halos. , keywords =

    work page

  28. [28]

    , keywords =

    Eccentricity evolution of PTA sources from cosmological initial conditions. , keywords =

    work page

  29. [29]

    , year = 1995, month = jan, volume =

    Inward Bound---The Search For Supermassive Black Holes In Galactic Nuclei. , year = 1995, month = jan, volume =

    work page 1995

  30. [30]

    , keywords =

    Supermassive Black Holes in Galactic Nuclei: Past, Present and Future Research. , keywords =

    work page

  31. [31]

    , keywords =

    Coevolution (Or Not) of Supermassive Black Holes and Host Galaxies. , keywords =

    work page

  32. [32]

    , keywords =

    Merger rates in hierarchical models of galaxy formation. , keywords =

    work page

  33. [33]

    , keywords =

    Another evolutionary correction to the luminosity of giant galaxies. , keywords =

    work page

  34. [34]

    , keywords =

    Cannibalism among the galaxies: dynamically produced evolution of cluster luminosity functions. , keywords =

    work page

  35. [35]

    , keywords =

    Theoretical Challenges in Galaxy Formation. , keywords =

    work page

  36. [36]

    , keywords =

    Massive black hole binaries in active galactic nuclei. , keywords =

    work page

  37. [37]

    Dynamical Friction. I. General Considerations: the Coefficient of Dynamical Friction. , year = 1943, month = mar, volume =

    work page 1943

  38. [38]

    , keywords =

    The effect of low-velocity, low-mass intruders (collisionless gas) on the dynamical evolution of a binary system. , keywords =

    work page

  39. [39]

    Hardening in a fixed stellar background

    The dynamical evolution of massive black hole binaries I. Hardening in a fixed stellar background. , keywords =

    work page

  40. [40]

    Physical Review , year = 1963, month = jul, volume =

    Gravitational Radiation from Point Masses in a Keplerian Orbit. Physical Review , year = 1963, month = jul, volume =

    work page 1963

  41. [41]

    Physical Review , year = 1964, month = nov, volume =

    Gravitational Radiation and the Motion of Two Point Masses. Physical Review , year = 1964, month = nov, volume =

    work page 1964

  42. [42]

    , keywords =

    Core condensation in heavy halos: a two-stage theory for galaxy formation and clustering. , keywords =

    work page

  43. [43]

    The second data release from the European Pulsar Timing Array. III. Search for gravitational wave signals. , keywords =

    work page

  44. [44]

    , keywords =

    Search for an Isotropic Gravitational-wave Background with the Parkes Pulsar Timing Array. , keywords =

    work page

  45. [45]

    Research in Astronomy and Astrophysics , keywords =

    Searching for the Nano-Hertz Stochastic Gravitational Wave Background with the Chinese Pulsar Timing Array Data Release I. Research in Astronomy and Astrophysics , keywords =

    work page

  46. [46]

    , keywords =

    The NANOGrav 15 yr Data Set: Evidence for a Gravitational-wave Background. , keywords =

    work page

  47. [47]

    , keywords =

    The NANOGrav 15 yr Data Set: Constraints on Supermassive Black Hole Binaries from the Gravitational-wave Background. , keywords =

    work page

  48. [48]

    arXiv e-prints , keywords =

    Multimessenger Probes of the Supermassive Black Hole Binary Population: The Role of Pulsar Timing Arrays. arXiv e-prints , keywords =

    work page

  49. [49]

    The second data release from the European Pulsar Timing Array. IV. Implications for massive black holes, dark matter, and the early Universe. , keywords =

    work page

  50. [50]

    , keywords =

    The EAGLE project: simulating the evolution and assembly of galaxies and their environments. , keywords =

    work page

  51. [51]

    , keywords =

    The multi-messenger view of pulsar timing array black holes with the Horizon-AGN simulation. , keywords =

    work page

  52. [52]

    , keywords =

    A new discrete dynamical friction estimator based on N-body simulations. , keywords =

    work page

  53. [53]

    , keywords =

    Massive Black Hole Merger Rates: The Effect of Kiloparsec Separation Wandering and Supernova Feedback. , keywords =

    work page

  54. [54]

    , keywords =

    Off the beaten path: a new approach to realistically model the orbital decay of supermassive black holes in galaxy formation simulations. , keywords =

    work page

  55. [55]

    A practical recipe for the evolution of massive black hole binaries in stellar environments

    Scattering experiments meet N-body - I. A practical recipe for the evolution of massive black hole binaries in stellar environments. , keywords =

    work page

  56. [56]

    , keywords =

    Handy Relation between Binary Black Hole Merger Times and Host Galaxy Properties. , keywords =

    work page

  57. [57]

    2013 , isbn =

    Dynamics and Evolution of Galactic Nuclei. 2013 , isbn =

    work page 2013

  58. [58]

    Journal of the Royal Statistical Society Series B: Statistical Methodology , volume=

    Regression shrinkage and selection via the lasso , author=. Journal of the Royal Statistical Society Series B: Statistical Methodology , volume=. 1996 , publisher=

    work page 1996

  59. [59]

    Machine Learning , volume=

    Random forests , author=. Machine Learning , volume=. 2001 , publisher=

    work page 2001

  60. [60]

    The Annals of Statistics , volume=

    Greedy function approximation: A gradient boosting machine , author=. The Annals of Statistics , volume=. 2001 , month=. doi:10.1214/aos/1013203451 , jstor=

    work page doi:10.1214/aos/1013203451 2001

  61. [61]

    The Annals of Mathematical Statistics , volume=

    A correspondence between Bayesian estimation on stochastic processes and smoothing by splines , author=. The Annals of Mathematical Statistics , volume=. 1970 , month=. doi:10.1214/aoms/1177697089 , issn=

    work page doi:10.1214/aoms/1177697089 1970

  62. [62]

    2006 , publisher=

    Gaussian processes for machine learning , author=. 2006 , publisher=. doi:10.7551/mitpress/3206.001.0001 , isbn=

    work page doi:10.7551/mitpress/3206.001.0001 2006

  63. [63]

    Nonlinear Component Analysis as a Kernel Eigenvalue Problem , year=

    Schölkopf, Bernhard and Smola, Alexander and Müller, Klaus-Robert , journal=. Nonlinear Component Analysis as a Kernel Eigenvalue Problem , year=

    work page

  64. [64]

    Frontiers in Artificial Intelligence , volume=

    An introduction to topological data analysis: fundamental and practical aspects for data scientists , author=. Frontiers in Artificial Intelligence , volume=. 2021 , month=. doi:10.3389/frai.2021.667963 , issn=

    work page doi:10.3389/frai.2021.667963 2021

  65. [65]

    , keywords =

    On the Eccentricity Distribution of Coalescing Black Hole Binaries Driven by the Kozai Mechanism in Globular Clusters. , keywords =

    work page

  66. [66]

    and Haberland, Matt and Reddy, Tyler and Cournapeau, David and Burovski, Evgeni and Peterson, Pearu and Weckesser, Warren and Bright, Jonathan and

    Virtanen, Pauli and Gommers, Ralf and Oliphant, Travis E. and Haberland, Matt and Reddy, Tyler and Cournapeau, David and Burovski, Evgeni and Peterson, Pearu and Weckesser, Warren and Bright, Jonathan and. Nature Methods , year =

    work page

  67. [67]

    Array Programming with NumPy

    Array programming with NumPy. , keywords =. doi:10.1038/s41586-020-2649-2 , archivePrefix =. 2006.10256 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1038/s41586-020-2649-2 2006

  68. [68]

    Computing in Science and Engineering , keywords =

    Matplotlib: A 2D Graphics Environment. Computing in Science and Engineering , keywords =

    work page

  69. [69]

    Computing in Science and Engineering , year = 2021, month = mar, volume =

    Jupyter: Thinking and Storytelling With Code and Data. Computing in Science and Engineering , year = 2021, month = mar, volume =

    work page 2021

  70. [70]

    1995 , publisher=

    Python reference manual , author=. 1995 , publisher=

    work page 1995

  71. [71]

    5 , author=

    pandas-dev/pandas: Pandas 1.0. 5 , author=. Zenodo , year=

    work page

  72. [72]

    the Journal of machine Learning research , volume=

    Scikit-learn: Machine learning in Python , author=. the Journal of machine Learning research , volume=. 2011 , publisher=

    work page 2011

  73. [73]

    Publications of the Astronomical Society of the Pacific , volume=

    emcee: the MCMC hammer , author=. Publications of the Astronomical Society of the Pacific , volume=. 2013 , publisher=

    work page 2013

  74. [74]

    , keywords =

    Realistic consecutive galaxy mergers form eccentric pulsar timing array sources. , keywords =. doi:10.1051/0004-6361/202556035 , archivePrefix =. 2506.16539 , primaryClass =

    work page doi:10.1051/0004-6361/202556035

  75. [75]

    Dynamical traction and black hole orbital migration

    Dynamical traction and black hole orbital migration: I. Angular momentum transfer and a fragmentation-driven instability. , keywords =. doi:10.1051/0004-6361/202453104 , archivePrefix =. 2507.09674 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1051/0004-6361/202453104

  76. [76]

    The Astrophysical Journal Letters , volume=

    Wandering black holes in bright disk galaxy halos , author=. The Astrophysical Journal Letters , volume=. 2010 , publisher=

    work page 2010

  77. [77]

    The Astrophysical Journal , volume=

    Growing massive black hole pairs in minor mergers of disk galaxies , author=. The Astrophysical Journal , volume=. 2011 , publisher=

    work page 2011

  78. [78]

    The role of gas in the merging of massive black holes in galactic nuclei. II. Black hole merging in a nuclear gas disk , author=. The Astrophysical Journal , volume=

    work page

  79. [79]

    Monthly Notices of the Royal Astronomical Society , volume=

    Stellar hardening of massive black hole binaries: the impact of the host rotation , author=. Monthly Notices of the Royal Astronomical Society , volume=. 2021 , publisher=

    work page 2021

  80. [80]

    Monthly Notices of the Royal Astronomical Society , volume=

    Supermassive black hole binaries in gaseous and stellar circumnuclear discs: orbital dynamics and gas accretion , author=. Monthly Notices of the Royal Astronomical Society , volume=. 2007 , publisher=

    work page 2007