Likelihood Methods for CMB Experiments

Gerbino, Martina; Lattanzi, M.; Migliaccio, M.; Pagano, L.; Salvati, L.; Colombo, L. P. L.; Gruppuso, A.; Natoli, P.; Polenta, G.

doi:10.3389/fphy.2020.00015

Cited by 18 publications

(15 citation statements)

References 101 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the present analysis, we follow the same overall strategy, although the lower level of systematics could have allowed a semi-analytical approach (see e.g. Mangilli et al (2015); Vanneste et al (2018); Hamimeche & Lewis (2008); Gerbino et al (2019)) which we leave to future analysis. With this method, we measure τ = 0.0566 +0.0053 −0.0062 at 68% C.L.…”

Section: Discussionmentioning

confidence: 99%

“…For details about validation and performances of the likelihood approximation seeGerbino et al (2019).5 The likelihood module is built within the clik infrastructure(Planck Collaboration XV 2014; Planck Collaboration ES 2013, 2015 and it is available on http://sroll20.ias.u-psud.fr or on https://web.fe.infn.it/∼pagano/low ell datasets/sroll2 6 http://cosmologist.info/cosmomc 7 http://camb.info…”

mentioning

confidence: 99%

See 1 more Smart Citation

Reionization optical depth determination from Planck HFI data with ten percent accuracy

Pagano

Delouis

Mottet

et al. 2020

A&A

View full text Add to dashboard Cite

We present an estimation of the reionization optical depth τ from an improved analysis of the High Frequency Instrument (HFI) data of Planck satellite. By using an improved version of the HFI map-making code, we greatly reduce the residual large scale contamination affecting the data, characterized, but not fully removed, in the Planck 2018 legacy release. This brings the dipole distortion systematic effect, contaminating the very low multipoles, below the noise level. On large scale polarization only data, we measure τ = 0.0566 +0.0053 −0.0062 at 68% C.L., reducing the Planck 2018 legacy release uncertainty by ∼ 40%. Within the ΛCDM model, in combination with the Planck large scale temperature likelihood, and the high-temperature and polarization likelihood, we measure τ = 0.059±0.006 at 68% C.L. which corresponds to a mid-point reionization redshift of z re = 8.14±0.61 at 68% C.L.. This estimation of the reionization optical depth with 10% accuracy is the strongest constraint to date.

show abstract

Section: Discussionmentioning

confidence: 99%

mentioning

confidence: 99%

Reionization optical depth determination from Planck HFI data with ten percent accuracy

Pagano

Delouis

Mottet

et al. 2020

A&A

View full text Add to dashboard Cite

show abstract

“…Moreover, being a real-space dataset, it is suitable for a number of studies that are not accessible for a spectrumbased likelihood (see, e.g., Planck Collaboration XXIII 2014;Planck Collaboration XVI 2016) and it is capable of exploring non-rotationally invariant cosmologies. For an exhaustive review of likelihood methods, see Gerbino et al (2020).…”

Section: Introductionmentioning

confidence: 99%

A novel CMB polarization likelihood package for large angular scales built from combined WMAP and Planck LFI legacy maps

Natale

Pagano

Lattanzi

et al. 2020

A&A

View full text Add to dashboard Cite

We present a cosmic microwave background (CMB) large-scale polarization dataset obtained by combining Wilkinson Microwave Anisotropy Probe (WMAP) in the K, Q, and V bands with the Planck 70 GHz maps. We employed the legacy frequency maps released by the WMAP and Planck collaborations and performed our own Galactic foreground mitigation technique, relying on Planck 353 GHz for polarized dust and on Planck 30 GHz and WMAP K for polarized synchrotron. We derived a single, optimally noise-weighted, low residual foreground map and the accompanying noise covariance matrix. These are shown through χ2 analysis to be robust over an ample collection of Galactic masks. We used this dataset, along with the Planck legacy Commander temperature solution, to build a pixel-based low-resolution CMB likelihood package, whose robustness we tested extensively with the aid of simulations, finding an excellent level of consistency. Using this likelihood package alone, we are able to constrain the optical depth to reionization, τ = 0.069−0.012+0.011 at 68% confidence level, on 54% of the sky. Adding the Planck high-ℓ temperature and polarization legacy likelihood, the Planck lensing likelihood, and BAO observations, we find τ = 0.0714−0.0096+0.0087 in a full ΛCDM exploration. The latter bounds are slightly less constraining than those obtained by employing the Planck High Frequency Instrument’s (HFI) CMB data for large-angle polarization, which only include EE correlations. Our bounds are based on a largely independent dataset that includes TE correlations. They are generally compatible with Planck HFI, but lean towards slightly higher values for τ. We have made the low-resolution Planck and WMAP joint dataset publicly available, along with the accompanying likelihood code.

show abstract

“…In particular, an entire slew of very diverse methods have been designed to produce estimates of the temperature or polarization power spectra or estimates of the cosmological parameters from a set of noisy maps of the sky. We can broadly classify these methods as pseudo-C l approaches, [1], [2], [3], [4], [5], see [6] for a review, maximum likelihood methods [7], [8] and Bayesian approaches using Monte-Carlo methods like Metropolis-Hastings [9], [10], [11], [12], Hamiltonian Monte-Carlo [13], [14], or Gibbs sampling [15], [16], [17], [18], [19]. Out of those, the pseudo-C l methods are very quick but require a careful design of a pseudo-likelihood for the data set of interest [6].…”

Section: Introductionmentioning

confidence: 99%

“…We can broadly classify these methods as pseudo-C l approaches, [1], [2], [3], [4], [5], see [6] for a review, maximum likelihood methods [7], [8] and Bayesian approaches using Monte-Carlo methods like Metropolis-Hastings [9], [10], [11], [12], Hamiltonian Monte-Carlo [13], [14], or Gibbs sampling [15], [16], [17], [18], [19]. Out of those, the pseudo-C l methods are very quick but require a careful design of a pseudo-likelihood for the data set of interest [6]. Monte-Carlo sampling techniques avoid this problem but are computationally demanding, in terms of both the time needed to draw a single sample and the number of required samples are concerned.…”

Section: Introductionmentioning

confidence: 99%

Improved Gibbs samplers for Cosmic Microwave Background power spectrum estimation

Ducrocq,

Chopin,

Errard

et al. 2021

Preprint

View full text Add to dashboard Cite

We study different variants of the Gibbs sampler algorithm from the perspective of their applicability to the estimation of power spectra of the cosmic microwave background (CMB) anisotropies. We focus on approaches which aim at reducing the cost of a computationally heavy constrained realization step and capitalize on the interweaving strategy to ensure that our algorithms mix well for high and low signal-to-noise ratio components. In particular, we propose an approach, which we refer to as Centered overrelax, which avoids the constraint realization step completely at the cost of additional, auxiliary variables and the need for overrelaxation. We demonstrate these variants and compare their merits on full and cut sky simulations, quantifying their performance in terms of an effective sample size (ESS) per second. We find that on nearly full-sky, satellite-like data, the proposed Gibbs sampler with overrelaxation performs between one and two orders of magnitude better than the usual Gibbs sampler potentially providing an interesting alternative to the currently favored approaches.

show abstract

Likelihood Methods for CMB Experiments

Cited by 18 publications

References 101 publications

Reionization optical depth determination from Planck HFI data with ten percent accuracy

Reionization optical depth determination from Planck HFI data with ten percent accuracy

A novel CMB polarization likelihood package for large angular scales built from combined WMAP and Planck LFI legacy maps

Improved Gibbs samplers for Cosmic Microwave Background power spectrum estimation

Contact Info

Product

Resources

About