pith. sign in

arxiv: 2502.11625 · v2 · submitted 2025-02-17 · 🌌 astro-ph.CO

Comparative Analysis of EMCEE, Gaussian Process, and Masked Autoregressive Flow in Constraining the Hubble Constant Using Cosmic Chronometers Dataset

Jing Niu , Jie-Feng Chen , Peng He , Tong-Jie Zhang , Jie Zhang This is my paper

Pith reviewed 2026-05-23 03:08 UTC · model grok-4.3

classification 🌌 astro-ph.CO
keywords Hubble constantcosmic chronometersEMCEEGaussian processmasked autoregressive flowsimulation testscosmological constraints
0
0 comments X

The pith

EMCEE recovers the Hubble constant with lower bias and better calibration than Gaussian processes or masked autoregressive flows in tests on mock cosmic chronometer data.

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

The paper compares three methods for turning H(z) measurements from cosmic chronometers into estimates of the Hubble constant. It first uses a delete-d jackknife procedure to measure how much each method's result shifts when individual data points are removed, finding GP most sensitive overall, followed by MAF, then EMCEE. It then runs simulation tests that generate many mock datasets with known true H0 under two different prescriptions, applies each method, and scores the recovered posteriors on bias, root-mean-square error, coverage of the true value, and log score. EMCEE ranks highest on these metrics, GP intermediate, and MAF lowest. Readers care because the same real dataset yields different H0 values depending on the method chosen, so identifying which approach is more reliable directly affects how chronometer data can be used to address the Hubble tension.

Core claim

When mock cosmic chronometer datasets are generated with a fixed input H0,EMCEE recovers posteriors with the smallest bias and RMSE, the coverage closest to the nominal 68 percent and 95 percent levels, and the highest log score. GP ranks in the middle on these quantities while MAF ranks lowest. GP is also the most sensitive to removal of single data points, MAF intermediate, and EMCEE least sensitive; EMCEE and GP respond more to high-redshift points while MAF responds more to low-redshift points. These rankings hold under both Lambda-CDM-based and GP-based mock-generation prescriptions.

What carries the argument

Monte Carlo delete-d jackknife (MCDJ) applied to assess point-wise sensitivity, together with simulation tests that compare recovered H0 posteriors to a known input truth via bias, RMSE, 68 percent and 95 percent coverage, and log score.

If this is right

  • EMCEE produces H0 constraints with lower bias and better-calibrated intervals than the other two methods across the tested mock ensembles.
  • Sensitivity rankings and redshift dependence differ systematically among the methods, so data-point influence is method-specific.
  • The performance ordering remains the same whether mocks are drawn from a Lambda-CDM model or from a GP model.
  • MAF shows the largest deviations from nominal coverage and the lowest log scores in the simulation tests.

Where Pith is reading between the lines

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

  • Analyses aiming to tighten constraints on the Hubble tension with chronometer data would gain by defaulting to EMCEE unless new evidence shows otherwise.
  • The redshift-dependent sensitivities could be used to design follow-up observations that target the redshift ranges where each method is least affected.
  • Extending the same mock-based ranking exercise to other cosmological probes might reveal whether EMCEE remains preferable outside the chronometer setting.

Load-bearing premise

The mock datasets generated under the Lambda-CDM and GP prescriptions accurately capture the statistical properties, error structure, and selection effects of real cosmic chronometer observations.

What would settle it

Repeating the simulation tests after changing the mock-generation prescription to include a different error distribution or redshift-dependent selection bias and finding that the performance ordering reverses would falsify the claim that EMCEE is the best performer.

Figures

Figures reproduced from arXiv: 2502.11625 by Jie-Feng Chen, Jie Zhang, Jing Niu, Peng He, Tong-Jie Zhang.

Figure 1
Figure 1. Figure 1: FIG. 1. Diagram of MRSA procedure. At the top, 26 out of 33 [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FIG. 2. The histogram of 1000 constrained H [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: FIG. 3. The histograms of the distribution of 1000 constrained H [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: FIG. 4. The 1 [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: FIG. 5. The simulated [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 9
Figure 9. Figure 9: FIG. 9. The simulated [PITH_FULL_IMAGE:figures/full_fig_p009_9.png] view at source ↗
Figure 8
Figure 8. Figure 8: FIG. 8. Distribution of [PITH_FULL_IMAGE:figures/full_fig_p009_8.png] view at source ↗
Figure 10
Figure 10. Figure 10: FIG. 10. The constrained 100 H [PITH_FULL_IMAGE:figures/full_fig_p009_10.png] view at source ↗
read the original abstract

The Hubble constant ($H_0$) is essential for understanding the universe's evolution. Different methods, such as Affine Invariant Markov chain Monte Carlo Ensemble sampler (EMCEE), Gaussian Process (GP), and Masked Autoregressive Flow (MAF), are used to constrain $H_0$ using $H(z)$ data. However, these methods produce varying $H_0$ values when applied to the same dataset. To investigate these differences, we compare the methods based on their sensitivity to individual data points and their performance in constraining $H_0$. We apply Monte Carlo delete-$d$ jackknife (MCDJ) to assess their sensitivity to individual data points. Our findings reveal that GP is more sensitive to individual data points than both MAF and EMCEE, with MAF being more sensitive than EMCEE. Sensitivity also depends on redshift: EMCEE and GP are more sensitive to $H(z)$ at higher redshifts, while MAF is more sensitive at lower redshifts. In simulation-based performance tests, we generate an ensemble of mock CC datasets with a fixed input truth $H_{0,\mathrm{true}}$, apply each method to recover $H_0$ posteriors, and summarise performance by comparing the recovered posterior to $H_{0,\mathrm{true}}$: (i) posterior central value accuracy (bias and RMSE), (ii) credible-interval calibration (68\% and 95\% coverage), and (iii) overall posterior quality (log score), under two simulation prescriptions ($\Lambda$CDM-based and GP-based). Overall, EMCEE performs best, GP is intermediate, and MAF performs worst across the performance metrics.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 3 minor

Summary. The paper compares three inference methods—EMCEE, Gaussian Process (GP), and Masked Autoregressive Flow (MAF)—for constraining the Hubble constant H0 from cosmic chronometer H(z) data. It applies Monte Carlo delete-d jackknife (MCDJ) to quantify sensitivity to individual data points (finding GP most sensitive, followed by MAF then EMCEE, with redshift dependence) and conducts simulation-based recovery tests on mock CC datasets generated under ΛCDM-based and GP-based prescriptions. Performance is ranked via bias, RMSE, 68%/95% coverage, and log score, with the headline result that EMCEE performs best, GP intermediate, and MAF worst.

Significance. If the performance ranking holds under more general mock-generation schemes, the work would provide a useful, quantitative guide for selecting inference engines when analyzing sparse H(z) datasets. The MCDJ sensitivity analysis and the multi-metric simulation protocol (bias/RMSE/coverage/log score) are concrete strengths that could be extended to other cosmological probes.

major comments (1)
  1. [simulation-based performance tests] The central performance ranking (EMCEE > GP > MAF) is obtained exclusively from recovery tests on mocks drawn from ΛCDM-based and GP-based prescriptions (abstract and simulation section). Because these prescriptions align with the parametric assumptions inside EMCEE and the kernel/smoothness assumptions inside GP, the design may systematically disadvantage the more flexible MAF; the manuscript does not report a cross-check using mocks generated by an independent process (e.g., bootstrap resampling of the real CC points with their covariance). This is load-bearing for the headline claim.
minor comments (3)
  1. [MCDJ analysis] The manuscript should specify the exact data-selection criteria, redshift range, and error model for the real CC sample used in the MCDJ analysis.
  2. [Methods] Hyperparameter choices for the GP kernel, MAF architecture (number of layers, hidden units, training schedule), and EMCEE settings should be stated explicitly, together with any sensitivity tests.
  3. [Results] Figure captions and table legends should clarify whether the reported coverage and log-score values are averaged over the ensemble of mocks or shown for representative realizations.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive review and for identifying a key aspect of our simulation protocol. We address the single major comment below.

read point-by-point responses

  1. Referee: [simulation-based performance tests] The central performance ranking (EMCEE > GP > MAF) is obtained exclusively from recovery tests on mocks drawn from ΛCDM-based and GP-based prescriptions (abstract and simulation section). Because these prescriptions align with the parametric assumptions inside EMCEE and the kernel/smoothness assumptions inside GP, the design may systematically disadvantage the more flexible MAF; the manuscript does not report a cross-check using mocks generated by an independent process (e.g., bootstrap resampling of the real CC points with their covariance). This is load-bearing for the headline claim.

    Authors: We thank the referee for this observation. The two mock-generation prescriptions were deliberately selected to span the modeling assumptions underlying EMCEE (parametric ΛCDM) and GP (non-parametric smoothness), thereby providing a controlled test in which the more flexible MAF is evaluated on data drawn from both classes of generative process. Nevertheless, we agree that an additional recovery test based on bootstrap resampling of the real CC points (respecting their reported covariances) would constitute an independent, data-driven check that does not presuppose any particular form for the underlying expansion history. We will therefore add this bootstrap-based recovery experiment, together with the corresponding performance metrics, to the revised manuscript. revision: yes

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The paper evaluates method performance exclusively via recovery on mock CC datasets that are generated with an externally fixed, known H0,true value under two stated simulation prescriptions. Recovered posteriors are scored against this independent truth using bias, RMSE, coverage, and log score. No quoted equation or procedure shows a fitted parameter being renamed as a prediction, a self-definitional loop, or a load-bearing self-citation that reduces the ranking to the input by construction. The simulation design and evaluation metrics remain external to the internal fitting steps of EMCEE, GP, and MAF, rendering the central claim self-contained.

Axiom & Free-Parameter Ledger

0 free parameters · 3 axioms · 0 invented entities

The analysis relies on standard assumptions of the three statistical methods and the realism of the mock data generation; no new free parameters, axioms beyond domain standards, or invented entities are introduced.

axioms (3)
  • standard math MCMC chains in EMCEE have converged to the target posterior
    Required for reliable posterior sampling with EMCEE.
  • domain assumption The chosen kernel and hyperparameters for the Gaussian Process adequately model the H(z) data covariance
    Central to GP performance and sensitivity results.
  • domain assumption The MAF architecture and training procedure produce well-calibrated densities for the parameter space
    Required for the flow-based inference to be comparable.

pith-pipeline@v0.9.0 · 5859 in / 1350 out tokens · 45520 ms · 2026-05-23T03:08:29.463276+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

54 extracted references · 54 canonical work pages · 28 internal anchors

  1. [1]

    In this paper, we adopt the flat ΛCDM cosmological model

    EMCEE To determine H 0 from CC data using EMCEE, it is essential to choose a cosmological model. In this paper, we adopt the flat ΛCDM cosmological model. The cor- responding Friedmann equation is given by: H(z) = H0 p ΩM(1 + z)3 + (1 − ΩM). (2) 3 Where H(z) represents the Hubble parameter, H 0 is the Hubble constant, Ω M denotes the matter density, and z...

    work page

  2. [2]

    GP is a powerful tool that can model the relationship in data using a joint Gaus- sian distribution

    GP H0 can also be determined by reconstructing the CC data using the Gaussian process, as outlined by Ras- mussen and Williams [6]. GP is a powerful tool that can model the relationship in data using a joint Gaus- sian distribution. It estimates values at new points with- out requiring additional parameters. In this study, we use a Gaussian process to est...

    work page

  3. [3]

    A reliable neural network estimator should balance both flexibility and tractability

    MAF The Hubble constant H 0 can be estimated using CC data through neural networks. A reliable neural network estimator should balance both flexibility and tractability. Two families that embody both properties are autore- gressive models [25] and normalizing flows [26]. In this study, we use MAF, proposed by Papamakarios et al. [8], to estimate H 0 from ...

    work page

  4. [4]

    • Step 3: Generate 1000 constrained values of H 0 with EMCEE, GP, and MAF respectively using the 1000 randomly selected H(z) datasets from Step 2

    This results in the creation of 1000 randomly selected datasets, labeled with an index i ranging from 0 to 999. • Step 3: Generate 1000 constrained values of H 0 with EMCEE, GP, and MAF respectively using the 1000 randomly selected H(z) datasets from Step 2. Each method generates 1000 values of H0, and these results are shown in Fig. 2. To evaluate the se...

    work page

  5. [5]

    We summarize these results in Table II

    we compare three metrics: (1) the absolute differ- ence between the mode of the H0 distribution and the H0 value constrained from the full 33 CC dataset, denoted as ∆ H0,mode−CC = |H0,mode − H0,CC|, (2) the absolute difference between the median of the H0 distribution and H0,CC, denoted as ∆H0,median−CC = |H0,median − H0,CC|, and (3) the range spanned by ...

    work page

  6. [6]

    The results are shown in Fig. 3. By analyzing Fig. 3, we can assess whether H(z) values from different redshift regions have varying impacts on the sensitivity of the constrained H0. The wider range of H0 within the 1σ (∆H0,1σ) and 2σ (∆H0,2σ) intervals in- dicates greater sensitivity of constrained H0 to individual H(z) data points in the corresponding r...

    work page

  7. [7]

    By combining these compo- nents, we can generate a simulated dataset consisting of 33 Hsim(zsim) data points based on the ΛCDM, as de- FIG

    At this stage, we have generated zsim, Hfid(zsim) and σsim(zsim) in Equation (20). By combining these compo- nents, we can generate a simulated dataset consisting of 33 Hsim(zsim) data points based on the ΛCDM, as de- FIG. 6. The simulated H(z) dataset based on the ΛCDM model using CC data. The simulated data points, Hsim(zsim), are shown as red triangles...

    work page

  8. [8]

    Planck 2018 results. VI. Cosmological parameters

    Planck Collaboration, N. Aghanim, Y. Akrami, M. Ash- down, J. Aumont, C. Baccigalupi, M. Ballardini, A. J. Banday, R. B. Barreiro, N. Bartolo, S. Basak, R. Battye, K. Benabed, J. P. Bernard, M. Bersanelli, P. Bielewicz, J. J. Bock, J. R. Bond, J. Borrill, F. R. Bouchet, F. Boulanger, M. Bucher, C. Burigana, R. C. Butler, E. Calabrese, J. F. Cardoso, J. Ca...

    work page internal anchor Pith review Pith/arXiv arXiv 2020

  9. [9]

    A. G. Riess, W. Yuan, L. M. Macri, D. Scolnic, D. Brout, S. Casertano, D. O. Jones, Y. Murakami, G. S. Anand, L. Breuval, T. G. Brink, A. V. Filippenko, S. Hoffmann, S. W. Jha, W. D’arcy Kenworthy, J. Mackenty, B. E. Stahl, and W. Zheng, Astrophys. J. L. 934, L7 (2022), arXiv:2112.04510 [astro-ph.CO]

    work page internal anchor Pith review Pith/arXiv arXiv 2022

  10. [10]

    Tensions between the Early and the Late Universe

    L. Verde, T. Treu, and A. G. Riess, Nature Astronomy 3, 891 (2019), arXiv:1907.10625 [astro-ph.CO]

    work page internal anchor Pith review Pith/arXiv arXiv 2019

  11. [11]

    A. G. Riess, Nature Reviews Physics 2, 10 (2020), arXiv:2001.03624 [astro-ph.CO]

    work page arXiv 2020

  12. [12]

    emcee: The MCMC Hammer

    D. Foreman-Mackey, D. W. Hogg, D. Lang, and J. Good- man, Publications of the Astronomical Society of the Pa- cific 125, 306 (2013), arXiv:1202.3665 [astro-ph.IM]

    work page internal anchor Pith review Pith/arXiv arXiv 2013

  13. [13]

    C. E. Rasmussen and C. K. I. Williams, Gaussian Pro- cesses for Machine Learning (2006)

    work page 2006

  14. [14]

    Reconstruction of dark energy and expansion dynamics using Gaussian processes

    M. Seikel, C. Clarkson, and M. Smith, JCAP 2012, 036 (2012), arXiv:1204.2832 [astro-ph.CO]

    work page internal anchor Pith review Pith/arXiv arXiv 2012

  15. [15]

    Masked Autoregressive Flow for Density Estimation

    G. Papamakarios, T. Pavlakou, and I. Murray, Proceed- ings of the 31st International Conference on Neural In- formation Processing Systems , arXiv:1705.07057 (2017), arXiv:1705.07057 [stat.ML]

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  16. [16]

    Cosmological Parameters from CMB Maps without Likelihood Approximation

    B. Racine, J. B. Jewell, H. K. Eriksen, and I. K. Wehus, Astrophys. J. 820, 31 (2016), arXiv:1512.06619 [astro- ph.CO]

    work page internal anchor Pith review Pith/arXiv arXiv 2016

  17. [17]

    Zhang, Y.-C

    H. Zhang, Y.-C. Wang, T.-J. Zhang, and T. Zhang, As- trophys. J. S. 266, 27 (2023), arXiv:2304.03911 [astro- ph.CO]

    work page arXiv 2023

  18. [18]

    Wang, Y.-B

    Y.-C. Wang, Y.-B. Xie, T.-J. Zhang, H.-C. Huang, T. Zhang, and K. Liu, Astrophys. J. S. 254, 43 (2021), arXiv:2005.10628 [astro-ph.CO]

    work page arXiv 2021

  19. [19]

    Constraining Cosmological Parameters Based on Relative Galaxy Ages

    R. Jimenez and A. Loeb, Astrophys. J. 573, 37 (2002), arXiv:astro-ph/0106145 [astro-ph]

    work page internal anchor Pith review Pith/arXiv arXiv 2002

  20. [20]

    Moresco, R

    M. Moresco, R. Jimenez, L. Verde, A. Cimatti, and L. Pozzetti, Astrophys. J. 898, 82 (2020), arXiv:2003.07362 [astro-ph.GA]

    work page arXiv 2020

  21. [21]

    K. Jiao, N. Borghi, M. Moresco, and T.-J. Zhang, As- 12 trophys. J. S. 265, 48 (2023), arXiv:2205.05701 [astro- ph.CO]

    work page arXiv 2023

  22. [22]

    Four New Observational $H(z)$ Data From Luminous Red Galaxies of Sloan Digital Sky Survey Data Release Seven

    C. Zhang, H. Zhang, S. Yuan, S. Liu, T.-J. Zhang, and Y.-C. Sun, Research in Astronomy and Astrophysics 14, 1221-1233 (2014), arXiv:1207.4541 [astro-ph.CO]

    work page internal anchor Pith review Pith/arXiv arXiv 2014

  23. [23]

    Constraints on the redshift dependence of the dark energy potential

    J. Simon, L. Verde, and R. Jimenez, Phys. Rev. D 71, 123001 (2005), arXiv:astro-ph/0412269 [astro-ph]

    work page internal anchor Pith review Pith/arXiv arXiv 2005

  24. [24]

    Improved constraints on the expansion rate of the Universe up to z~1.1 from the spectroscopic evolution of cosmic chronometers

    M. Moresco, A. Cimatti, R. Jimenez, L. Pozzetti, G. Zamorani, M. Bolzonella, J. Dunlop, F. Lamareille, M. Mignoli, H. Pearce, P. Rosati, D. Stern, L. Verde, E. Zucca, C. M. Carollo, T. Contini, J. P. Kneib, O. Le F` evre, S. J. Lilly, V. Mainieri, A. Renzini, M. Scodeggio, I. Balestra, R. Gobat, R. McLure, S. Bardelli, A. Bon- giorno, K. Caputi, O. Cuccia...

    work page internal anchor Pith review Pith/arXiv arXiv 2012

  25. [25]

    A 6% measurement of the Hubble parameter at $z\sim0.45$: direct evidence of the epoch of cosmic re-acceleration

    M. Moresco, L. Pozzetti, A. Cimatti, R. Jimenez, C. Maraston, L. Verde, D. Thomas, A. Citro, R. To- jeiro, and D. Wilkinson, JCAP 2016, 014 (2016), arXiv:1601.01701 [astro-ph.CO]

    work page internal anchor Pith review Pith/arXiv arXiv 2016

  26. [26]

    A. L. Ratsimbazafy, S. I. Loubser, S. M. Craw- ford, C. M. Cress, B. A. Bassett, R. C. Nichol, and P. V¨ ais¨ anen, Mon. Not. Roy. Astron. Soc. 467, 3239 (2017), arXiv:1702.00418 [astro-ph.CO]

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  27. [27]

    Cosmic Chronometers: Constraining the Equation of State of Dark Energy. I: H(z) Measurements

    D. Stern, R. Jimenez, L. Verde, M. Kamionkowski, and S. A. Stanford, JCAP 2010, 008 (2010), arXiv:0907.3149 [astro-ph.CO]

    work page internal anchor Pith review Pith/arXiv arXiv 2010

  28. [28]

    Tomasetti, M

    E. Tomasetti, M. Moresco, N. Borghi, K. Jiao, A. Cimatti, L. Pozzetti, A. C. Carnall, R. J. McLure, and L. Pentericci, Astron. Astrophys. 679, A96 (2023), arXiv:2305.16387 [astro-ph.CO]

    work page arXiv 2023

  29. [29]

    Raising the bar: new constraints on the Hubble parameter with cosmic chronometers at z$\sim$2

    M. Moresco, Mon. Not. Roy. Astron. Soc. 450, L16 (2015), arXiv:1503.01116 [astro-ph.CO]

    work page internal anchor Pith review Pith/arXiv arXiv 2015

  30. [30]

    Power of Observational Hubble Parameter Data: a Figure of Merit Exploration

    C. Ma and T.-J. Zhang, Astrophys. J. 730, 74 (2011), arXiv:1007.3787 [astro-ph.CO]

    work page internal anchor Pith review Pith/arXiv arXiv 2011

  31. [31]

    W. Sun, K. Jiao, and T.-J. Zhang, Astrophys. J. 915, 123 (2021), arXiv:2105.12618 [astro-ph.CO]

    work page arXiv 2021

  32. [32]

    Neural Autoregressive Distribution Estimation

    B. Uria, M.-A. Cˆ ot´ e, K. Gregor, I. Murray, and H. Larochelle, The Journal of Machine Learning Research , arXiv:1605.02226 (2016), arXiv:1605.02226 [cs.LG]

    work page internal anchor Pith review Pith/arXiv arXiv 2016

  33. [33]

    Jimenez Rezende and S

    D. Jimenez Rezende and S. Mohamed, Proceedings of the 32nd International Conference on Machine Learning (2015)

    work page 2015

  34. [34]

    MADE: Masked Autoencoder for Distribution Estimation

    M. Germain, K. Gregor, I. Murray, and H. Larochelle, arXiv e-prints , arXiv:1502.03509 (2015), arXiv:1502.03509 [cs.LG]

    work page internal anchor Pith review Pith/arXiv arXiv 2015

  35. [35]

    Variational Inference with Normalizing Flows

    D. Jimenez Rezende and S. Mohamed, arXiv e-prints , arXiv:1505.05770 (2015), arXiv:1505.05770 [stat.ML]

    work page internal anchor Pith review Pith/arXiv arXiv 2015

  36. [36]

    Chen, Y.-C

    J.-F. Chen, Y.-C. Wang, T. Zhang, and T.-J. Zhang, Phys. Rev. D 107, 063517 (2023), arXiv:2211.05064 [astro-ph.CO]

    work page arXiv 2023

  37. [37]

    A. Endo, E. van Leeuwen, and M. Baguelin, Epidemics 29, 100363 (2019)

    work page 2019

  38. [38]

    Y. Li, S. Rao, A. Hassaine, R. Ramakrishnan, D. Canoy, G. Salimi-Khorshidi, M. Mamouei, T. Lukasiewicz, and K. Rahimi, Scientific Reports 11, 20685 (2021)

    work page 2021

  39. [39]

    Thomas, Computational Economics 60, 451 (2022)

    L. Thomas, Computational Economics 60, 451 (2022)

    work page 2022

  40. [40]

    Rostam, R

    M. Rostam, R. Nagamune, and V. Grebenyuk, Journal of Process Control 92, 149 (2020)

    work page 2020

  41. [41]

    Astropy Collaboration, A. M. Price-Whelan, B. M. Sip˝ ocz, H. M. G¨ unther, P. L. Lim, S. M. Crawford, S. Conseil, D. L. Shupe, M. W. Craig, N. Dencheva, A. Ginsburg, J. T. VanderPlas, L. D. Bradley, D. P´ erez- Su´ arez, M. de Val-Borro, T. L. Aldcroft, K. L. Cruz, T. P. Robitaille, E. J. Tollerud, C. Ardelean, T. Babej, Y. P. Bach, M. Bachetti, A. V. Ba...

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  42. [42]

    Astropy Collaboration, T. P. Robitaille, E. J. Tollerud, P. Greenfield, M. Droettboom, E. Bray, T. Ald- croft, M. Davis, A. Ginsburg, A. M. Price-Whelan, W. E. Kerzendorf, A. Conley, N. Crighton, K. Barbary, D. Muna, H. Ferguson, F. Grollier, M. M. Parikh, P. H. Nair, H. M. Unther, C. Deil, J. Woillez, S. Conseil, R. Kramer, J. E. H. Turner, L. Singer, R....

    work page internal anchor Pith review Pith/arXiv arXiv 2013

  43. [43]

    Bertin and S

    E. Bertin and S. Arnouts, Astron. Astrophys. Suppl. Ser. 117, 393 (1996)

    work page 1996

  44. [44]

    Quantifying the Observational Effort Required for the Radial Velocity Characterization of TESS Planets

    R. Cloutier, R. Doyon, F. Bouchy, and G. H´ ebrard, As- tron. J. 156, 82 (2018), arXiv:1807.01263 [astro-ph.EP]

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  45. [45]

    X-ray Scattering Echoes and Ghost Halos from the Intergalactic Medium: Relation to the nature of AGN variability

    L. Corrales, Astrophys. J. 805, 23 (2015), arXiv:1503.01475 [astro-ph.HE]

    work page internal anchor Pith review Pith/arXiv arXiv 2015

  46. [46]

    G. J. Ferland, R. L. Porter, P. A. M. van Hoof, R. J. R. Williams, N. P. Abel, M. L. Lykins, G. Shaw, W. J. Henney, and P. C. Stancil, Revista Mexicana de As- tronom´ ıa y Astrof´ ısica49, 137 (2013), arXiv:1302.4485 13 [astro-ph.GA]

    work page internal anchor Pith review Pith/arXiv arXiv 2013

  47. [47]

    R. J. Hanisch and C. D. Biemesderfer, in Bulletin of the American Astronomical Society (1989) p. 780

    work page 1989

  48. [48]

    Lamport, LaTeX: A Document Preparation System , 2nd ed

    L. Lamport, LaTeX: A Document Preparation System , 2nd ed. (Addison-Wesley Professional, 1994)

    work page 1994

  49. [49]

    L. Li, J. Zhang, H. Peter, L. P. Chitta, J. Su, H. Song, C. Xia, and Y. Hou, Astrophys. J. 868, L33 (2018), arXiv:1811.08553 [astro-ph.SR]

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  50. [50]

    Nominal values for selected solar and planetary quantities: IAU 2015 Resolution B3

    A. Prˇ sa, P. Harmanec, G. Torres, E. Mamajek, M. As- plund, N. Capitaine, J. Christensen-Dalsgaard, ´E. De- pagne, M. Haberreiter, and S. Hekker, Astron. J. 152, 41 (2016), arXiv:1605.09788 [astro-ph.SR]

    work page internal anchor Pith review Pith/arXiv arXiv 2016

  51. [51]

    G. J. Schwarz, J.-U. Ness, J. P. Osborne, K. L. Page, P. A. Evans, A. P. Beardmore, F. M. Walter, L. A. Hel- ton, C. E. Woodward, M. Bode, S. Starrfield, and J. J. Drake, Astrophys. J. S. 197, 31 (2011), arXiv:1110.6224 [astro-ph.SR]

    work page internal anchor Pith review Pith/arXiv arXiv 2011

  52. [52]

    F. P. A. Vogt, M. A. Dopita, L. J. Kewley, R. S. Suther- land, J. Scharw¨ achter, H. M. Basurah, A. Ali, and M. A. Amer, Astrophys. J. 793, 127 (2014), arXiv:1406.5186 [astro-ph.GA]

    work page internal anchor Pith review Pith/arXiv arXiv 2014

  53. [53]

    J. Niu, Y. Chen, and T.-J. Zhang, arXiv e-prints , arXiv:2305.04752 (2023), arXiv:2305.04752 [astro- ph.CO]

    work page arXiv 2023

  54. [54]

    Niu and T.-J

    J. Niu and T.-J. Zhang, Physics of the Dark Universe 39, 101147 (2023), arXiv:2204.10597 [astro-ph.CO]

    work page arXiv 2023