Pith. Sign in

REVIEW

Cosmological parameters from CMB and other data: a Monte-Carlo approach

PaperPDF from arXiv

data files

ReviewThe objections and verdict, as JSON Paper breakdownThe authors' claims, conditions and measurements, as JSON Signed receiptProof this review has not been altered
X Bluesky LinkedIn Reddit Hacker News
Antony Lewis, Sarah Bridle

Astrophysics · Submitted 24 May 2002 · 58 references

Pith's verdict: The methods paper that became the field's plumbing — released code, released chains, and unusually honest diagnostics; the m_nu limit is the one number to quote with care.

T0 · AI review

How this review was made: An AI referee (selfhosted/deepseek-v4.1-flash) read the paper on October 29, 2026 and answered a fixed list of questions about it. T0 is the starting level every paper gets. The levels describe how deeply Pith has checked the paper, never the authors.

  1. T0 AI review An AI read the paper and answered a fixed list of questions.
  2. T1 Audit Automatic checks that the references and files the paper relies on exist and match.
  3. T2 Compute The paper's code or calculations re-run with the same results.
  4. T3 Formal statement The claims written out in a formal math language a computer can read.
  5. T4 Proof Those claims proved, and the proof checked by computer.

How Pith checks papers →

· 2 major objections · 9 minor · Ask about this paper · The compiled record · Read the full review →
Challenge this review Comment (0) Read on arXiv ↗
Cite ▾
arXiv IDastro-ph/0205436 (version 3)
Pith IDAIGBOIFS, Pith's permanent record of this paper and its review
CitationAntony Lewis, Sarah Bridle. "Cosmological parameters from CMB and other data: a Monte-Carlo approach." arXiv:astro-ph/0205436 (2002). https://pith.science/paper/AIGBOIFS
Cite this reviewBibTeX for Pith's review of this paper
Claim paper

Why Pith says this

Highlight any text to comment on it.

Referee report

How Pith checks papers → · Full audit of this paper →

The paper

The paper claims

A random-walk sampler turns cosmological data into six numbers, and the chains are handed out for reuse.

The reading

This paper asks whether a random-walk sampler — Markov Chain Monte Carlo — can replace grid-based parameter fitting in cosmology, and answers yes by extracting constraints from the then-new CMB band powers combined with the Hubble Key Project, the 2dF galaxy survey, Type Ia supernovae and big-bang nucleosynthesis. The payoff is a compact set of numbers for a flat six-parameter model: $h = 0.67$ with a $4$ per cent error, $n_s = 0.98$, $t_0 = 14.1$ Gyr, and a matter density $0.30$ with a $0.04$ error, with the dark-energy density filling the rest. The paper's most distinctive physics claim is that the CMB alone, without any galaxy data, fixes the amplitude of matter clustering, through two parameter combinations the authors isolate by principal component analysis. It then shows that the extra freedom of nine and eleven parameters — massive neutrinos, a varying dark-energy equation of state, tensor modes, spatial curvature — does not improve the fit, and quotes upper limits rather than detections. The methodological contribution is the release of the chains themselves: new data can be folded in later by re-weighting existing samples rather than recomputing anything.

What carries the argument

The engine is the Metropolis-Hastings algorithm: a chain steps through parameter space, proposing a new point from a fixed proposal density and accepting it with probability equal to the ratio of the posterior densities, which guarantees that the long-run density of chain positions is the posterior itself. Two refinements do most of the practical work. First, the proposal density is split into 'fast' parameters (those that only change the normalization or shape of an already-computed spectrum) and 'slow' parameters (those that require re-running the Boltzmann code), so a step touches only one to three parameters at a time and the acceptance rate stays near 50 per cent. Second, importance sampling re-weights an existing set of samples by the ratio of a new posterior to the old one, so that adding a Hubble prior, adding the 2dF data, or correcting an approximate power-spectrum generator costs one pass over the samples rather than a new chain. Analytic marginalization over unknown calibrations and over the supernova intrinsic magnitude and the galaxy bias amplitude removes those nuisance parameters from the sampling entirely.

Extended reading notes

Core claim

The central claim is that a joint Monte Carlo analysis of CMB band powers and four other data sets yields the constraints tabulated for 6, 9 and 11 parameter models, and that the CMB alone already determines the matter clustering amplitude. On the CMB-only side the authors find two well-determined orthogonal combinations, $v_1 = 0.113$ and $v_2 = 0.70$, which together encode $v_1 = 0.113$ in the combination $v_1 = 0.113$ of $v_1 = 0.113$ — concretely, $v_1 = 0.113$ in the combination $v_1 = 0.113$ and $v_2 = 0.70$ in the combination $v_2 = 0.70$, giving $v_1 = 0.113$ and $v_2 = 0.70$. Restated plainly: the CMB fixes $v_1 = 0.113$ in the combination $v_1 = 0.113$ and $v_2 = 0.70$ in the combination $v_2 = 0.70$, so that the matter clustering amplitude is pinned once the optical depth and the Hubble parameter are specified. With all data combined the paper reports $m_ν < 0.3$ eV and $w < -0.75$ at 95 per cent confidence for the nine-parameter flat model, $r_{10} < 0.7$ for the tensor-to-scalar ratio, and $Ω_K = 0.00 ± 0.02$ for curvature; in the eleven-parameter non-flat model these weaken to $m_ν < 0.4$ eV and $w < -0.69$. The authors are explicit that these are upper limits, not detections: there is no evidence in the data for massive neutrinos, for $w ≠ -1$, or for tensor modes.

Load-bearing premise

The neutrino mass limit rests on the assumption that the 2dF galaxy power spectrum over $0.02 < k/(h$ Mpc$^{-1}) < 0.15$ is directly proportional to the linear matter power spectrum today, meaning that galaxy bias, evolution and redshift-space distortions are all scale independent over that range.

Editorial extensions

If this is right

  • The CMB alone supplies a normalization of the matter power spectrum, so the clustering amplitude can be quoted without borrowing a prior from galaxy surveys or cluster counts.
  • The neutrino mass limit of about 0.3 eV at 95 per cent confidence is set mainly by the shape of the 2dF power spectrum between $0.02$ and $0.15\,h\,\mathrm{Mpc}^{-1}$, where massive neutrinos damp small-scale power.
  • Curvature is constrained to $Ω_K = 0.00 ± 0.02$ once all data are combined, and the dark-energy equation of state is bounded to $w < -0.75$, both consistent with a flat universe and a cosmological constant.
  • Adding three or five extra parameters does not improve the average fit: the mean likelihood ratios for the 6, 9 and 11 parameter models are 1 : 0.4 : 0.3, so the simplest model remains the best bet on average.
  • Because the chains are released, a new data set can be incorporated by re-weighting the published samples instead of re-running the cosmology.

Reading between the lines

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

  • If the released chains are reused as intended, the expensive part of cosmological parameter estimation shifts from computing theory spectra to agreeing on priors and likelihoods, which makes prior choice the dominant source of disagreement between groups rather than the physics code.
  • The paper's own demonstration that a flat prior on $h$ versus a flat prior on $Ω_Λ$ moves the 95 per cent upper limit on $h$ from 0.89 to 0.59 suggests that any future 'tension' between cosmological data sets should be quoted alongside the base-parameter choice that generated it.
  • The neutrino bound reported here sits close to the minimum total mass that oscillation experiments require for three neutrino species, so a modest improvement in the small-scale matter power measurement would turn this upper limit into a genuine detection rather than a constraint.
  • The mean-likelihood-versus-marginalized-posterior diagnostic introduced in the appendices is a cheap way to spot when a parameter's preferred value is being driven by the volume of parameter space rather than by the data, and it could be applied routinely to any posterior sampled by chains.
Original abstract

We present a fast Markov Chain Monte-Carlo exploration of cosmological parameter space. We perform a joint analysis of results from recent CMB experiments and provide parameter constraints, including sigma_8, from the CMB independent of other data. We next combine data from the CMB, HST Key Project, 2dF galaxy redshift survey, supernovae Ia and big-bang nucleosynthesis. The Monte Carlo method allows the rapid investigation of a large number of parameters, and we present results from 6 and 9 parameter analyses of flat models, and an 11 parameter analysis of non-flat models. Our results include constraints on the neutrino mass (m_nu < 0.3eV), equation of state of the dark energy, and the tensor amplitude, as well as demonstrating the effect of additional parameters on the base parameter constraints. In a series of appendices we describe the many uses of importance sampling, including computing results from new data and accuracy correction of results generated from an approximate method. We also discuss the different ways of converting parameter samples to parameter constraints, the effect of the prior, assess the goodness of fit and consistency, and describe the use of analytic marginalization over normalization parameters.

The compiled record

The paper's own sentences, arranged, kept beside the review and separate from it. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.

Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-09-23: 1 request, 5,724 input tokens.

What the abstract itself states, as probabilities: claims first 0.03 · states limitation 0.17 · human participants 0.01 · controlled comparison 0.04 · randomized assignment 0.01 · states quantitative result 0.98 · compares with prior results 0.24 · mentions code or data availability 0.04.

The abstract, sentence by sentence

arXiv listing abstract v3. Results and the main claim are underlined; .

We present a fast Markov Chain Monte-Carlo exploration of cosmological parameter space. We perform a joint analysis of results from recent CMB experiments and provide parameter constraints, including sigma_8, from the CMB independent of other data. We next combine data from the CMB, HST Key Project, 2dF galaxy redshift survey, supernovae Ia and big-bang nucleosynthesis. The Monte Carlo method allows the rapid investigation of a large number of parameters, and we present results from 6 and 9 parameter analyses of flat models, and an 11 parameter analysis of non-flat models. Our results include constraints on the neutrino mass (m_nu < 0.3eV), equation of state of the dark energy, and the tensor amplitude, as well as demonstrating the effect of additional parameters on the base parameter constraints. In a series of appendices we describe the many uses of importance sampling, including computing results from new data and accuracy correction of results generated from an approximate method. We also discuss the different ways of converting parameter samples to parameter constraints, the effect of the prior, assess the goodness of fit and consistency, and describe the use of analytic marginalization over normalization parameters.

background contribution method result interpretation main claim

What the classifier reports for each sentence. The number is the probability it gave the label.
#sentencerolethe authors' own workstates the main conclusionstrength
s0 We present a fast Markov Chain Monte-Carlo exploration of cosmological parameter space. contribution 0.56 0.96 0.07 asserted 0.99
s1 We perform a joint analysis of results from recent CMB experiments and provide parameter constraints, including sigma_8, from the CMB independent of other data. contribution 0.70 0.95 0.20 asserted 1.00
s2 We next combine data from the CMB, HST Key Project, 2dF galaxy redshift survey, supernovae Ia and big-bang nucleosynthesis. method 0.69 0.87 0.05 asserted 0.93
s3 The Monte Carlo method allows the rapid investigation of a large number of parameters, and we present results from 6 and 9 parameter analyses of flat models, and an 11 parameter analysis of non-flat models. method 0.66 0.93 0.09 asserted 0.99
s4 Our results include constraints on the neutrino mass (m_nu < 0.3eV), equation of state of the dark energy, and the tensor amplitude, as well as demonstrating the effect of additional parameters on the base parameter constraints. result 1.00 0.95 0.28 asserted 1.00
s5 In a series of appendices we describe the many uses of importance sampling, including computing results from new data and accuracy correction of results generated from an approximate method. method 0.72 0.76 0.03 asserted 0.96
s6 We also discuss the different ways of converting parameter samples to parameter constraints, the effect of the prior, assess the goodness of fit and consistency, and describe the use of analytic marginalization over normalization parameters. method 0.82 0.74 0.04 asserted 0.93

What kind of paper the classifier reports this is: method or tool 0.64 · evidence from analysis of existing data 0.91.

What was examined, and what was not

sources read
  • arXiv listing abstract v3
not examined
  • full text
  • figures and tables
  • references and citing sentences
  • relations to other papers
levels run
  • P abstract
not run
  • S sections
  • T body sentences
  • C citations
  • R cross-paper relations

Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. Every number in the measurements table is the authors', checked by code. Record as JSON

Figures
Figure 1. FIG. 1: The CMB temperature anisotropy band-power data used
Figure 3. FIG. 3: Posterior constraints for 9-parameter flat models us
Figure 4. FIG. 4: All-data posterior constraints for flat inflationary
Figure 5. FIG. 5: Posterior constraints for 11-parameter non-flat mod
Figure 6. FIG. 6: Posterior constraints for 11-parameter non-flat mod
Figure 7. FIG. 7: Parameter constrains from the CMB alone with flat prio

Showing 6 of the paper's 7 figures. See all figures in the PDF ↗ Figures reproduced from arXiv:astro-ph/0205436 by the authors.

References
How later papers used it

1,129 later papers cite it

Evidence: substantial. 1,120 of the citing papers share no author with it, and 1,005 of those carry a confident label from Pith's citation classifier.

Pith read 40 of the 1,299 citing sentences, choosing only those whose citation markers point at this paper alone, and wrote the paraphrases from them. The summary, the counts, the evidence level, the links and the quotes come from code; the summary counts the papers listed below. A later paper is listed under a label when Pith's citation classifier gives one of its sentences that label at probability 0.8 or more (0.95 for support), Pith's reading finds that the citing paper itself did it, and a second check confirms the paraphrase against the sentence. This section lists uses only. It does not list criticism or disagreement, so their absence here says nothing about how the paper has been received. Papers that share an author with this one are listed apart, under "By the same authors". A quote is the citing authors' own sentence, shown when their paper's licence is CC BY, CC BY-SA or CC0. JSON

Uses its method or tool 34 of 1,002 papers with this label
How the literature uses it

799 arXiv papers cite this in 917 sentences

Each citing sentence is taken from the citing paper's LaTeX source and typed by Pith's classifier. A label appears at probability 0.8 or more (0.95 for support), a contest at 0.7 or more; the quoted words are the citing authors'. How each label was checked · JSON

Papers citing it · 88 sentences without a confident label

Uses its method or tool 718 papers
Uses its data or values 5 papers
  • Planck 2018 data [ref] used together with BICEP/Keck 2018 data [ref] give the following contraints on $n_s$ and $r$ at $k_CMB=0.05 Mpc^-1$ in the $ΛCDM+r$ cosmology [ref] [equation] The two-dimensional likelihood[The authors are very grateful to Daniela Paoletti for providing the original MCMC chains produced using the sampler [ref] , which Planck and BICEP/Keck 2018 constraints on the standard $Λ$CDM + $r$ model in [ref] were based on.

    Novel CMB constraints on the $\alpha$ parameter in alpha-attractor models · CMB constraints on $α$ and the duration of reheating · 1.00

  • Consistency with observational data (the temperature, polarization, and lensing of the Planck 2018 likelihood data [ref] , baryon acoustic oscillations (BAO) data from 6dF, SDSS DR7, and SDSS DR12 galaxy surveys [ref] , and the Type Ia Supernovae Pantheon data [ref] ) was determined using Cobaya [ref] along with a Monte Carlo Markov Chain sampler [ref] , which used the fast-dragging procedure described in [ref] .

    Observational constraints on inflection point quintessence with a cubic potential · Observational limits · 0.88

  • In light of the observed level of B-modes, as well as the presence of a slight tension at large scales between the data and the theory, we performed an analysis of the BICEP2 data [ref] in combination with the Planck [We use the low $ℓ$ WMAP polarization data contained in Planck likelihood code.] public likelihood [ref] and a modified version of the public cosmology package Cosmomc [ref] .

    Alleviating the tension at low multipole through Axion Monodromy · Results · 0.90

Compares with it 6 papers
  • In <Ref> we report the approximate expected total runtime of compared to the implementation of the MCMC sampler [ref] and the nested sampler [ref] (via its interface), neglecting the overhead of MCMC and as in this case it is tiny [ref] .

    Fast and robust Bayesian Inference using Gaussian Processes with GPry · Modelling the marginalized posterior · 0.86

  • This was also observed by the Planck collaboration [ref] , although their values of $Y_p$ are somewhat lower.[We found results closer to the Planck values when using CosmoMC/CAMB [ref] rather than Monte Python/CLASS in the analysis with varying $$ and $Y_p$.

    Phases of New Physics in the CMB · Planck 2015 Results · 0.91

  • In contrast, the Planck collaboration reports $100θ_MC$, which is given by Eq. (6) in [ref] , and corresponds to the standard choice in [ref] .

    Five parameters are all you need (in $\Lambda$CDM) · MCMC inference · 0.94

Cites it as background 19 papers

Discussion

Disagree with the review, or know something it missed? Leave a comment, highlight any passage above to comment on it, or challenge this review formally.