Search

Non-Gaussian statistics of the projected weak lensing field are powerful estimators that can outperform the constraining power of the two-point functions in inferring cosmological parameters. This is because these estimators extract the non-Gaussian information contained in the small scales. However, fully leveraging the statistical precision of such estimators is hampered by theoretical uncertainties, such as those arising from baryonic physics. Moreover, as non-Gaussian estimators mix different scales, there exists no natural cut-off scale below which baryonic feedback can be completely removed. We therefore present a Bayesian solution for accounting for baryonic feedback uncertainty in weak lensing non-Gaussianity inference. Our solution implements Bayesian model averaging (BMA), a statistical framework that accounts for model uncertainty and combines the strengths of different models to produce more robust and reliable parameter inferences. We demonstrate the effectiveness of this approach in a Stage IV convergence peak counts analysis, including three baryonic feedback models. We find that the resulting BMA posterior distribution safeguards parameter inference against biases due to baryonic feedback, and therefore provides a robust framework for obtaining accurate cosmological constraints at Stage IV precision under model uncertainty scenarios., Comment: 9 pages, 3 figures. Accepted for publication in MNRAS

The ratio of Bayesian evidences is a popular tool in cosmology to compare different models. There are however several issues with this method: Bayes' ratio depends on the prior even in the limit of non-informative priors, and Jeffrey's scale, used to assess the test, is arbitrary. Moreover, the standard use of Bayes' ratio is often criticized for being unable to reject models. In this paper, we address these shortcoming by promoting evidences and evidence ratios to frequentist statistics and deriving their sampling distributions. By comparing the evidence ratios to their sampling distributions, poor fitting models can now be rejected. Our method additionally does not depend on the prior in the limit of very weak priors, thereby safeguarding the experimenter against premature rejection of a theory with a uninformative prior, and replaces the arbitrary Jeffrey's scale by probability thresholds for rejection. We provide analytical solutions for some simplified cases (Gaussian data, linear parameters, and nested models), and we apply the method to cosmological supernovae Ia data. We dub our method the FB method, for Frequentist-Bayesian., Comment: 20 pages; v2: added some citations and small edits; v3: minor edits to match version accepted in Phys. Rev. D

Simulating the evolution of the gravitational N-body problem becomes extremely computationally expensive as N increases since the problem complexity scales quadratically with the number of bodies. We study the use of Artificial Neural Networks (ANNs) to replace expensive parts of the integration of planetary systems. Neural networks that include physical knowledge have grown in popularity in the last few years, although few attempts have been made to use them to speed up the simulation of the motion of celestial bodies. We study the advantages and limitations of using Hamiltonian Neural Networks to replace computationally expensive parts of the numerical simulation. We compare the results of the numerical integration of a planetary system with asteroids with those obtained by a Hamiltonian Neural Network and a conventional Deep Neural Network, with special attention to understanding the challenges of this problem. Due to the non-linear nature of the gravitational equations of motion, errors in the integration propagate. To increase the robustness of a method that uses neural networks, we propose a hybrid integrator that evaluates the prediction of the network and replaces it with the numerical solution if considered inaccurate. Hamiltonian Neural Networks can make predictions that resemble the behavior of symplectic integrators but are challenging to train and in our case fail when the inputs differ ~7 orders of magnitude. In contrast, Deep Neural Networks are easy to train but fail to conserve energy, leading to fast divergence from the reference solution. The hybrid integrator designed to include the neural networks increases the reliability of the method and prevents large energy errors without increasing the computing cost significantly. For this problem, the use of neural networks results in faster simulations when the number of asteroids is >70., Comment: Accepted for publication in the Journal of Computational Physics

We develop extreme data compression for use in Bayesian model comparison via the MOPED algorithm, as well as more general score compression. We find that Bayes factors from data compressed with the MOPED algorithm are identical to those from their uncompressed datasets when the models are linear and the errors Gaussian. In other nonlinear cases, whether nested or not, we find negligible differences in the Bayes factors, and show this explicitly for the Pantheon-SH0ES supernova dataset. We also investigate the sampling properties of the Bayesian Evidence as a frequentist statistic, and find that extreme data compression reduces the sampling variance of the Evidence, but has no impact on the sampling distribution of Bayes factors. Since model comparison can be a very computationally-intensive task, MOPED extreme data compression may present significant advantages in computational time., Comment: 15 pages, 5 figures. Invited paper for JCAP 20th anniversary edition

In order to understand grain-surface chemistry, one must have a good understanding of the reaction rate parameters. For diffusion-based reactions, these parameters are binding energies of the reacting species. However, attempts to estimate these values from grain-surface abundances using Bayesian inference are inhibited by a lack of enough sufficiently constraining data. In this work, we use the Massive Optimised Parameter Estimation and Data (MOPED) compression algorithm to determine which species should be prioritised for future ice observations to better constrain molecular binding energies. Using the results from this algorithm, we make recommendations for which species future observations should focus on., Comment: 10 pages, 4 figures, accepted for publication in MNRAS

Neural nets have become popular to accelerate parameter inferences, especially for the upcoming generation of galaxy surveys in cosmology. As neural nets are approximative by nature, a recurrent question has been how to propagate the neural net's approximation error, in order to avoid biases in the parameter inference. We present a Bayesian solution to propagating a neural net's approximation error and thereby debiasing parameter inference. We exploit that a neural net reports its approximation errors during the validation phase. We capture the thus reported approximation errors via the highest-order summary statistics, allowing us to eliminate the neural net's bias during inference, and propagating its uncertainties. We demonstrate that our method is quickly implemented and successfully infers parameters even for strongly biased neural nets. In summary, our method provides the missing element to judge the accuracy of a posterior if it cannot be computed based on an infinitely accurately theory code., Comment: 7 pages, 9 figures. Accepted by the Open Journal of Astrophysics

We present a variational-Bayes solution to compute non-Gaussian posteriors from extremely expensive likelihoods. Our approach is an alternative for parameter inference when MCMC sampling is numerically prohibitive or conceptually unfeasible. For example, when either the likelihood or the theoretical model cannot be evaluated at arbitrary parameter values, but only previously selected values, then traditional MCMC sampling is impossible, whereas our variational-Bayes solution still succeeds in estimating the full posterior. In cosmology, this occurs e.g. when the parametric model is based on costly simulations that were run for previously selected input parameters. We demonstrate the applicability of our posterior construction on the KiDS-450 weak lensing analysis, where we reconstruct the original KiDS MCMC posterior at 0.6% of its former numerical posterior evaluations. The reduction in numerical cost implies that systematic effects which formerly exhausted the numerical budget could now be included., Comment: Submitted to MNRAS, 10 pages, 5 figures

The velocity-space distribution of the solar neighbourhood stars shows complex substructures. Most of the previous studies use static potentials to investigate their origins. Instead we use a self-consistent $N$-body model of the Milky Way, whose potential is asymmetric and evolves with time. In this paper, we quantitatively evaluate the similarities of the velocity-space distributions in the $N$-body model and that of the solar neighbourhood, using Kullback-Leibler divergence (KLD). The KLD analysis shows the time evolution and spatial variation of the velocity-space distribution. The KLD fluctuates with time, which indicates the velocity-space distribution at a fixed position is not always similar to that of the solar neighbourhood. Some positions show velocity-space distributions with small KLDs (high similarities) more frequently than others. One of them locates at $(R,\phi)=(8.2\;\mathrm{kpc}, 30^{\circ})$, where $R$ and $\phi$ are the distance from the galactic centre and the angle with respect to the bar's major axis, respectively. The detection frequency is higher in the inter-arm regions than in the arm regions. In the velocity maps with small KLDs, we identify the velocity-space substructures, which consist of particles trapped in bar resonances. The bar resonances have significant impact on the stellar velocity-space distribution even though the galactic potential is not static., Comment: 9 pages, 11 figures. Accepted by MNRAS

Observational astrophysics consists of making inferences about the Universe by comparing data and models. The credible intervals placed on model parameters are often as important as the maximum a posteriori probability values, as the intervals indicate concordance or discordance between models and with measurements from other data. Intermediate statistics (e.g. the power spectrum) are usually measured and inferences made by fitting models to these rather than the raw data, assuming that the likelihood for these statistics has multivariate Gaussian form. The covariance matrix used to calculate the likelihood is often estimated from simulations, such that it is itself a random variable. This is a standard problem in Bayesian statistics, which requires a prior to be placed on the true model parameters and covariance matrix, influencing the joint posterior distribution. As an alternative to the commonly-used Independence-Jeffreys prior, we introduce a prior that leads to a posterior that has approximately frequentist matching coverage. This is achieved by matching the covariance of the posterior to that of the distribution of true values of the parameters around the maximum likelihood values in repeated trials, under certain assumptions. Using this prior, credible intervals derived from a Bayesian analysis can be interpreted approximately as confidence intervals, containing the truth a certain proportion of the time for repeated trials. Linking frequentist and Bayesian approaches that have previously appeared in the astronomical literature, this offers a consistent and conservative approach for credible intervals quoted on model parameters for problems where the covariance matrix is itself an estimate., Comment: 15 pages, 6 figures, version accepted by MNRAS

Structure formation in our Universe creates non-Gaussian random fields that will soon be observed over almost the entire sky by the Euclid satellite, the Vera-Rubin observatory, and the Square Kilometre Array. An unsolved problem is how to analyze best such non-Gaussian fields, e.g. to infer the physical laws that created them. This problem could be solved if a parametric non-Gaussian sampling distribution for such fields were known, as this distribution could serve as likelihood during inference. We therefore create a sampling distribution for non-Gaussian random fields. Our approach is capable of handling strong non-Gaussianity, while perturbative approaches such as the Edgeworth expansion cannot. To imitate cosmological structure formation, we enforce our fields to be (i) statistically isotropic, (ii) statistically homogeneous, and (iii) statistically independent at large distances. We generate such fields via a Monte Carlo Markov Chain technique and find that even strong non-Gaussianity is not necessarily visible to the human eye. We also find that sampled marginals for pixel pairs have an almost generic Gauss-like appearance, even if the joint distribution of all pixels is markedly non-Gaussian. This apparent Gaussianity is a consequence of the high dimensionality of random fields. We conclude that vast amounts of non-Gaussian information can be hidden in random fields that appear nearly Gaussian in simple tests, and that it would be short-sighted not to try and extract it., Comment: Accepted by the Open Journal of Astrophysics

Upcoming galaxy surveys will allow us to probe the growth of the cosmic large-scale structure with improved sensitivity compared to current missions, and will also map larger areas of the sky. This means that in addition to the increased precision in observations, future surveys will also access the ultra-large-scale regime, where commonly neglected effects such as lensing, redshift-space distortions and relativistic corrections become important for calculating correlation functions of galaxy positions. At the same time, several approximations usually made in these calculations, such as the Limber approximation, break down at those scales. The need to abandon these approximations and simplifying assumptions at large scales creates severe issues for parameter estimation methods. On the one hand, exact calculations of theoretical angular power spectra become computationally expensive, and the need to perform them thousands of times to reconstruct posterior probability distributions for cosmological parameters makes the approach unfeasible. On the other hand, neglecting relativistic effects and relying on approximations may significantly bias the estimates of cosmological parameters. In this work, we quantify this bias and investigate how an incomplete modelling of various effects on ultra-large scales could lead to false detections of new physics beyond the standard $\Lambda$CDM model. Furthermore, we propose a simple debiasing method that allows us to recover true cosmologies without running the full parameter estimation pipeline with exact theoretical calculations. This method can therefore provide a fast way of obtaining accurate values of cosmological parameters and estimates of exact posterior probability distributions from ultra-large-scale observations., Comment: 16 pages, 8 figures. Version matching the one published by MNRAS

Non-Gaussian (NG) statistics of the thermal Sunyaev-Zeldovich (tSZ) effect carry significant information which is not contained in the power spectrum. Here, we perform a joint Fisher analysis of the tSZ power spectrum and bispectrum to verify how much the full bispectrum can contribute to improve parameter constraints. We go beyond similar studies of this kind in several respects: first of all, we include the complete power spectrum and bispectrum (auto- and cross-) covariance in the analysis, computing all NG contributions; furthermore we consider a multi-component foreground scenario and model the effects of component separation in the forecasts; finally, we consider an extended set of both cosmological and intra-cluster medium parameters. We show that the tSZ bispectrum is very efficient at breaking parameter degeneracies, making it able to produce even stronger cosmological constraints than the tSZ power spectrum: e.g. the standard deviation on $\sigma_8$ shrinks from $\sigma^\text{PS}(\sigma_8)=0.35$ to $\sigma^\text{BS}(\sigma_8)=0.065$ when we consider a multi-parameter analysis. We find that this is mostly due to the different response of separate triangle types (e.g. equilateral and squeezed) to changes in model parameters. While weak, this shape dependence is clearly non-negligible for cosmological parameters, and it is even stronger, as expected, for intra-cluster medium parameters., Comment: 25 pages + 14 pages of appendices and references. 13 figures, 9 tables. Submitted to JCAP

Weak lensing data follow a naturally skewed distribution, implying the data vector most likely yielded from a survey will systematically fall below its mean. Although this effect is qualitatively known from CMB-analyses, correctly accounting for it in weak lensing is challenging, as a direct transfer of the CMB results is quantitatively incorrect. While a previous study (Sellentin et al. 2018) focused on the magnitude of this bias, we here focus on the frequency of this bias, its scaling with redshift, and its impact on the signal-to-noise of a survey. Filtering weak lensing data with COSEBIs, we show that weak lensing likelihoods are skewed up until $\ell \approx 100$, whereas CMB-likelihoods Gaussianize already at $\ell \approx 20$. While COSEBI-compressed data on KiDS- and DES-like redshift- and angular ranges follow Gaussian distributions, we detect skewness at 6$\sigma$ significance for half of a Euclid- or LSST-like data set, caused by the wider coverage and deeper reach of these surveys. Computing the signal-to-noise ratio per data point, we show that precisely the data points of highest signal-to-noise are the most biased. Over all redshifts, this bias affects at least 10% of a survey's total signal-to-noise, at high redshifts up to 25%. The bias is accordingly expected to impact parameter inference. The bias can be handled by developing non-Gaussian likelihoods. Otherwise, it could be reduced by removing the data points of highest signal-to-noise., Comment: Accepted by the Open Journal of Astrophysics

Stream stars removed by tides from their progenitor satellite galaxy or globular cluster act as a group of test particles on neighboring orbits, probing the gravitational field of the Milky Way. While constraints from individual streams have been shown to be susceptible to biases, combining several streams from orbits with various distances reduces these biases. We fit a common gravitational potential to multiple stellar streams simultaneously by maximizing the clustering of the stream stars in action space. We apply this technique to members of the GD-1, Pal 5, Orphan and Helmi streams, exploiting both the individual and combined data sets. We describe the Galactic potential with a St\"ackel model, and vary up to five parameters simultaneously. We find that we can only constrain the enclosed mass, and that the strongest constraints come from the GD-1, Pal 5 and Orphan streams whose combined data set yields $M(< 20\ \mathrm{kpc}) = 2.96^{+0.25}_{-0.26} \times 10^{11} \ M_{\odot}$. When including the Helmi stream in the data set, the mass uncertainty increases to $M(< 20\ \mathrm{kpc}) = 3.12^{+3.21}_{-0.46} \times 10^{11} \ M_{\odot}$., Comment: 25 pages, 15 figures. Accepted for publication in MNRAS

Bringing a high-dimensional dataset into science-ready shape is a formidable challenge that often necessitates data compression. Compression has accordingly become a key consideration for contemporary cosmology, affecting public data releases, and reanalyses searching for new physics. However, data compression optimized for a particular model can suppress signs of new physics, or even remove them altogether. We therefore provide a solution for exploring new physics \emph{during} data compression. In particular, we store additional agnostic compressed data points, selected to enable precise constraints of non-standard physics at a later date. Our procedure is based on the maximal compression of the MOPED algorithm, which optimally filters the data with respect to a baseline model. We select additional filters, based on a generalised principal component analysis, which are carefully constructed to scout for new physics at high precision and speed. We refer to the augmented set of filters as MOPED-PC. They enable an analytic computation of Bayesian evidences that may indicate the presence of new physics, and fast analytic estimates of best-fitting parameters when adopting a specific non-standard theory, without further expensive MCMC analysis. As there may be large numbers of non-standard theories, the speed of the method becomes essential. Should no new physics be found, then our approach preserves the precision of the standard parameters. As a result, we achieve very rapid and maximally precise constraints of standard and non-standard physics, with a technique that scales well to large dimensional datasets., Comment: 12 pages, 12 figures. Accepted for publication in MNRAS

Gaia Data Release 2 revealed detailed structures of nearby stars in phase space. These include the Hercules stream, whose origin is still debated. Most of the previous numerical studies conjectured that the observed structures originate from orbits in resonance with the bar, based on static potential models for the Milky Way. We, in contrast, approach the problem via a self-consistent, dynamic, and morphologically well-resolved model, namely a full $N$-body simulation of the Milky Way. Our simulation comprises about 5.1 billion particles in the galactic stellar bulge, bar, disk, and dark-matter halo and is evolved to 10 Gyr. Our model's disk component is composed of 200 million particles, and its simulation snapshots are stored every 10 Myr, enabling us to resolve and classify resonant orbits of representative samples of stars. After choosing the Sun's position in the simulation, we compare the distribution of stars in its neighborhood with Gaia's astrometric data, thereby establishing the role of identified resonantly trapped stars in the formation of Hercules-like structures. From our orbital spectral-analysis we identify multiple, especially higher order resonances. Our results suggest that the Hercules stream is dominated by the 4:1 and 5:1 outer Lindblad and corotation resonances. In total, this yields a trimodal structure of the Hercules stream. From the relation between resonances and ridges in phase space, our model favored a slow pattern speed of the Milky-Way bar (40--45 $\mathrm{km \; s^{-1} \; kpc^{-1}}$)., Comment: 11 pages, 9 figures, MNRAS accepted

The Tri-Agency Cosmological Simulations (TACS) Task Force was formed when Program Managers from the Department of Energy (DOE), the National Aeronautics and Space Administration (NASA), and the National Science Foundation (NSF) expressed an interest in receiving input into the cosmological simulations landscape related to the upcoming DOE/NSF Vera Rubin Observatory (Rubin), NASA/ESA's Euclid, and NASA's Wide Field Infrared Survey Telescope (WFIRST). The Co-Chairs of TACS, Katrin Heitmann and Alina Kiessling, invited community scientists from the USA and Europe who are each subject matter experts and are also members of one or more of the surveys to contribute. The following report represents the input from TACS that was delivered to the Agencies in December 2018., Comment: 36 pages, 3 figures. Delivered to NASA, NSF, and DOE in Dec 2018

This paper presents a joint blinding and deblinding strategy for inference of physical laws from astronomical data. The strategy allows for up to three blinding stages, where the data may be blinded, the computations of theoretical physics may be blinded, and --assuming Gaussianly distributed data-- the covariance matrix may be blinded. We found covariance blinding to be particularly effective, as it enables the blinder to determine close to exactly where the blinded posterior will peak. Accordingly, we present an algorithm which induces posterior shifts in predetermined directions by hiding untraceable biases in a covariance matrix. The associated deblinding takes the form of a numerically lightweight post-processing step, where the blinded posterior is multiplied with deblinding weights. We illustrate the blinding strategy for cosmic shear from KiDS-450, and show that even though there is no direct evidence of the KiDS-450 covariance matrix being biased, the famous cosmic shear tension with Planck could easily be induced by a mischaracterization of correlations between $\xi_-$ at the highest redshift and all lower redshifts. The blinding algorithm illustrates the increasing importance of accurate uncertainty assessment in astronomical inferences, as otherwise involuntary blinding through biases occurs.

We develop a fully non-invasive use of machine learning in order to enable open research on Euclid-sized data sets. Our algorithm leaves complete control over theory and data analysis, unlike many black-box like uses of machine learning. Focusing on a `3x2 analysis' which combines cosmic shear, galaxy clustering and tangential shear at a Euclid-like sky coverage, we arrange a total of 348000 data points into data matrices whose structure permits not only an easy prediction by neural nets, but it additionally permits the essential removal from the data of patterns which the neural nets could not `understand'. The latter provides an often lacking mechanism to control and debias the inference of physics. The theoretical backbone to our neural net training can be any conventional (deterministic) theory code, where we chose CLASS. After training, we infer the seven parameters of a $w$CDM cosmology by Monte Carlo Markov sampling posteriors at Euclid-like precision within a day. We publicly provide the neural nets which memorise and output all 3x2 power spectra at a Euclid-like sky coverage and redshift binning., Comment: Original version of the authors, as accepted by the referee(s)

When a posterior peaks in unexpected regions of parameter space, new physics has either been discovered, or a bias has not been identified yet. To tell these two cases apart is of paramount importance. We therefore present a method to indicate and mitigate unrecognized biases: Our method runs any pipeline with possibly unknown biases on both simulations and real data. It computes the coverage probability of posteriors, which measures whether posterior volume is a faithful representation of probability or not. If found to be necessary, the posterior is then corrected. This is a non-parametric debiasing procedure which complies with objective Bayesian inference. We use the method to debias inference with approximate covariance matrices and redshift uncertainties. We demonstrate why approximate covariance matrices bias physical constraints, and how this bias can be mitigated. We show that for a Euclid-like survey, if a traditional likelihood exists, then 25 end-to-end simulations suffice to guarantee that the figure of merit deteriorates maximally by 22 percent, or by 10 percent for 225 simulations. Thus, even a pessimistic analysis of Euclid-like data will still constitute an 25-fold increase in precision on the dark energy parameters in comparison to the state of the art (2018) set by KiDS and DES. We provide a public code of our method.

Given the precision of current neutrino data, priors still impact noticeably the constraints on neutrino masses and their hierarchy. To avoid our understanding of neutrinos being driven by prior assumptions, we construct a prior that is mathematically minimally informative. Using the constructed uninformative prior, we find that the normal hierarchy is favoured but with inconclusive posterior odds of 5.1:1. Better data is hence needed before the neutrino masses and their hierarchy can be well constrained. We find that the next decade of cosmological data should provide conclusive evidence if the normal hierarchy with negligible minimum mass is correct, and if the uncertainty in the sum of neutrino masses drops below 0.025 eV. On the other hand, if neutrinos obey the inverted hierarchy, achieving strong evidence will be difficult with the same uncertainties. Our uninformative prior was constructed from principles of the Objective Bayesian approach. The prior is called a reference prior and is minimally informative in the specific sense that the information gain after collection of data is maximised. The prior is computed for the combination of neutrino oscillation data and cosmological data and still applies if the data improve., Comment: 15 pages. Minor changes to match accepted version in JCAP

We investigate the corrections which relativistic light-cone computations induce on the correlation of the tangential shear with galaxy number counts, also known as galaxy-galaxy lensing. The standard-approach to galaxy-galaxy lensing treats the number density of sources in a foreground bin as observable, whereas it is in reality unobservable due to the presence of relativistic corrections. We find that already in the redshift range covered by the DES first year data, these currently neglected relativistic terms lead to a systematic correction of up to 50% in the density-shear correlation function for the highest redshift bins. This correction is dominated by the the fact that a redshift bin of number counts does not only lens sources in a background bin, but is itself again lensed by all masses between the observer and the counted source population. Relativistic corrections are currently ignored in the standard galaxy-galaxy analyses, and the additional lensing of a counted source populations is only included in the error budget (via the covariance matrix). At increasingly higher redshifts and larger scales, these relativistic and lensing corrections become however increasingly more important, and we here argue that it is then more efficient, and also cleaner, to account for these corrections in the density-shear correlations., Comment: version published in JCAP

We derive the essentials of the skewed weak lensing likelihood via a simple Hierarchical Model. Our likelihood passes four objective and cosmology-independent tests which a standard Gaussian likelihood fails. We demonstrate that sound weak lensing analyses are naturally biased low, and this does not indicate any new physics such as deviations from $\Lambda$CDM. Mathematically, the biases arise because noisy two-point functions follow skewed distributions. This form of bias is already known from CMB analyses, where the low multipoles have asymmetric error bars. Weak lensing is more strongly affected by this asymmetry as galaxies form a discrete set of shear tracer particles, in contrast to a smooth shear field. We demonstrate that the biases can be up to 30 percent of the standard deviation per data point, dependent on the properties of the weak lensing survey. Our likelihood provides a versatile framework with which to address this bias in future weak lensing analyses., Comment: Submitted to MNRAS

Non-linear gravitational collapse introduces non-Gaussian statistics into the matter fields of the late Universe. As the large-scale structure is the target of current and future observational campaigns, one would ideally like to have the full probability density function of these non-Gaussian fields. The only viable way we see to achieve this analytically, at least approximately and in the near future, is via the Edgeworth expansion. We hence rederive this expansion for Fourier modes of non-Gaussian fields and then continue by putting it into a wider statistical context than previously done. We show that in its original form, the Edgeworth expansion only works if the non-Gaussian signal is averaged away. This is counterproductive, since we target the parameter-dependent non-Gaussianities as a signal of interest. We hence alter the analysis at the decisive step and now provide a roadmap towards a controlled and unadulterated analysis of non-Gaussianities in structure formation (with the Edgeworth expansion). Our central result is that, although the Edgeworth expansion has pathological properties, these can be predicted and avoided in a careful manner. We also show that, despite the non-Gaussianity coupling all modes, the Edgeworth series may be applied to any desired subset of modes, since this is equivalent (to the level of the approximation) to marginalising over the exlcuded modes. In this first paper of a series, we restrict ourselves to the sampling properties of the Edgeworth expansion, i.e.~how faithfully it reproduces the distribution of non-Gaussian data. A follow-up paper will detail its Bayesian use, when parameters are to be inferred., Comment: 25 pages, 7 figures

We derive an exact expression for the correlation function in redshift shells including all the relativistic contributions. This expression, which does not rely on the distant-observer or flat-sky approximation, is valid at all scales and includes both local relativistic corrections and integrated contributions, like gravitational lensing. We present two methods to calculate this correlation function, one which makes use of the angular power spectrum C_ell(z1,z2) and a second method which evades the costly calculations of the angular power spectra. The correlation function is then used to define the power spectrum as its Fourier transform. In this work theoretical aspects of this procedure are presented, together with quantitative examples. In particular, we show that gravitational lensing modifies the multipoles of the correlation function and of the power spectrum by a few percent at redshift z=1 and by up to 30% and more at z=2. We also point out that large-scale relativistic effects and wide-angle corrections generate contributions of the same order of magnitude and have consequently to be treated in conjunction. These corrections are particularly important at small redshift, z=0.1, where they can reach 10%. This means in particular that a flat-sky treatment of relativistic effects, using for example the power spectrum, is not consistent., Comment: 49 pages, 19 figures. v2: minor corrections, references added, typos corrected in eqs. (2.3)-(2.6). Matches version published in JCAP

We show how the massive data compression algorithm MOPED can be used to reduce, by orders of magnitude, the number of simulated datasets that are required to estimate the covariance matrix required for the analysis of gaussian-distributed data. This is relevant when the covariance matrix cannot be calculated directly. The compression is especially valuable when the covariance matrix varies with the model parameters. In this case, it may be prohibitively expensive to run enough simulations to estimate the full covariance matrix throughout the parameter space. This compression may be particularly valuable for the next-generation of weak lensing surveys, such as proposed for Euclid and LSST, for which the number of summary data (such as band power or shear correlation estimates) is very large, $\sim 10^4$, due to the large number of tomographic redshift bins that the data will be divided into. In the pessimistic case where the covariance matrix is estimated separately for all points in an MCMC analysis, this may require an unfeasible $10^9$ simulations. We show here that MOPED can reduce this number by a factor of 1000, or a factor of $\sim 10^6$ if some regularity in the covariance matrix is assumed, reducing the number of simulations required to a manageable $10^3$, making an otherwise intractable analysis feasible., Comment: 8 pages. Accepted by MNRAS

We investigate whether a Gaussian likelihood, as routinely assumed in the analysis of cosmological data, is supported by simulated survey data. We define test statistics, based on a novel method that first destroys Gaussian correlations in a dataset, and then measures the non-Gaussian correlations that remain. This procedure flags pairs of datapoints which depend on each other in a non-Gaussian fashion, and thereby identifies where the assumption of a Gaussian likelihood breaks down. Using this diagnostic, we find that non-Gaussian correlations in the CFHTLenS cosmic shear correlation functions are significant. With a simple exclusion of the most contaminated datapoints, the posterior for $s_8$ is shifted without broadening, but we find no significant reduction in the tension with $s_8$ derived from Planck Cosmic Microwave Background data. However, we also show that the one-point distributions of the correlation statistics are noticeably skewed, such that sound weak lensing data sets are intrinsically likely to lead to a systematically low lensing amplitude being inferred. The detected non-Gaussianities get larger with increasing angular scale such that for future wide-angle surveys such as Euclid or LSST, with their very small statistical errors, the large-scale modes are expected to be increasingly affected. The shifts in posteriors may then not be negligible and we recommend that these diagnostic tests be run as part of future analyses., Comment: Replacement to match accepted MNRAS version

In this paper, we present a method for computing the marginal likelihood, also known as the model likelihood or Bayesian evidence, from Markov Chain Monte Carlo (MCMC), or other sampled posterior distributions. In order to do this, one needs to be able to estimate the density of points in parameter space, and this can be challenging in high numbers of dimensions. Here we present a Bayesian analysis, where we obtain the posterior for the marginal likelihood, using $k$th nearest-neighbour distances in parameter space, using the Mahalanobis distance metric, under the assumption that the points in the chain (thinned if required) are independent. We generalise the algorithm to apply to importance-sampled chains, where each point is assigned a weight. We illustrate this with an idealised posterior of known form with an analytic marginal likelihood, and show that for chains of length $\sim 10^5$ points, the technique is effective for parameter spaces with up to $\sim 20$ dimensions. We also argue that $k=1$ is the optimal choice, and discuss failure modes for the algorithm. In a companion paper (Heavens et al. 2017) we apply the technique to the main MCMC chains from the 2015 Planck analysis of cosmic background radiation data, to infer that quantitatively the simplest 6-parameter flat $\Lambda$CDM standard model of cosmology is preferred over all extensions considered.

We compute the Bayesian Evidence for models considered in the main analysis of Planck cosmic microwave background data. By utilising carefully-defined nearest-neighbour distances in parameter space, we reuse the Monte Carlo Markov Chains already produced for parameter inference to compute Bayes factors $B$ for many different model-dataset combinations. Standard 6-parameter flat $\Lambda$CDM model is favoured over all other models considered, with curvature being mildly favoured only when CMB lensing is not included. Many alternative models are strongly disfavoured by the data, including primordial correlated isocurvature models ($\ln B=-7.8$), non-zero scalar-to-tensor ratio ($\ln B=-4.3$), running of the spectral index ($\ln B = -4.7$), curvature ($\ln B=-3.6$), non-standard numbers of neutrinos ($\ln B=-3.1$), non-standard neutrino masses ($\ln B=-3.2$), non-standard lensing potential ($\ln B=-4.6$), evolving dark energy ($\ln B=-3.2$), sterile neutrinos ($\ln B=-6.9$), and extra sterile neutrinos with a non-zero scalar-to-tensor ratio ($\ln B=-10.8$). Other models are less strongly disfavoured with respect to flat $\Lambda$CDM. As with all analyses based on Bayesian Evidence, the final numbers depend on the widths of the parameter priors. We adopt the priors used in the Planck analysis, while performing a prior sensitivity analysis. Our quantitative conclusion is that extensions beyond the standard cosmological model are disfavoured by Planck data. Only when newer Hubble constant measurements are included does $\Lambda$CDM become disfavoured, and only mildly, compared with a dynamical dark energy model ($\ln B\sim +2$)., Comment: 5 pages. Accepted for publication in PRL. Effect of inclusion of recent H0 measurements is added

Parameter inference with an estimated covariance matrix systematically loses information due to the remaining uncertainty of the covariance matrix. Here, we quantify this loss of precision and develop a framework to hypothetically restore it, which allows to judge how far away a given analysis is from the ideal case of a known covariance matrix. We point out that it is insufficient to estimate this loss by debiasing a Fisher matrix as previously done, due to a fundamental inequality that describes how biases arise in non-linear functions. We therefore develop direct estimators for parameter credibility contours and the figure of merit. We apply our results to DES Science Verification weak lensing data, detecting a 10% loss of information that increases their credibility contours. No significant loss of information is found for KiDS. For a Euclid-like survey, with about 10 nuisance parameters we find that 2900 simulations are sufficient to limit the systematically lost information to 1%, with an additional uncertainty of about 2%. Without any nuisance parameters 1900 simulations are sufficient to only lose 1% of information. We also derive an estimator for the Fisher matrix of the unknown true covariance matrix, two estimators of its inverse with different physical meanings, and an estimator for the optimally achievable figure of merit. The formalism here quantifies the gains to be made by running more simulated datasets, allowing decisions to be made about numbers of simulations in an informed way., Comment: Figure 6 holds for all surveys; replacement to match published version

When inferring parameters from a Gaussian-distributed data set by computing a likelihood, a covariance matrix is needed that describes the data errors and their correlations. If the covariance matrix is not known a priori, it may be estimated and thereby becomes a random object with some intrinsic uncertainty itself. We show how to infer parameters in the presence of such an estimated covariance matrix, by marginalising over the true covariance matrix, conditioned on its estimated value. This leads to a likelihood function that is no longer Gaussian, but rather an adapted version of a multivariate t-distribution, which has the same numerical complexity as the multivariate Gaussian. As expected, marginalisation over the true covariance matrix improves inference when compared with Hartlap et al.'s method, which uses an unbiased estimate of the inverse covariance matrix but still assumes that the likelihood is Gaussian., Comment: Matches accepted MNRAS letter

Assuming a Euclid-like weak lensing data set, we compare different methods of dealing with its inherent parameter degeneracies. Including priors into a data analysis can mask the information content of a given data set alone. However, since the information content of a data set is usually estimated with the Fisher matrix, priors are added in order to enforce an approximately Gaussian likelihood. Here, we compare priorless forecasts to more conventional forecasts that use priors. We find strongly non-Gaussian likelihoods for 2d-weak lensing if no priors are used, which we approximate with the DALI-expansion. Without priors, the Fisher matrix of the 2d-weak lensing likelihood includes unphysical values of $\Omega_m$ and $h$, since it does not capture the shape of the likelihood well. The Cramer-Rao inequality then does not need to apply. We find that DALI and Monte Carlo Markov Chains predict the presence of a dark energy with high significance, whereas a Fisher forecast of the same data set also allows decelerated expansion. We also find that a 2d-weak lensing analysis provides a sharp lower limit on the Hubble constant of $h > 0.4$, even if the equation of state of dark energy is jointly constrained by the data. This is not predicted by the Fisher matrix and usually masked in other works by a sharp prior on $h$. Additionally, we find that DALI estimates Figures of Merit in the presence of non-Gaussianities better than the Fisher matrix. We additionally demonstrate how DALI allows switching to a Hamiltonian Monte Carlo sampling of a highly curved likelihood with acceptance rates of $\approx 0.5$, an effective covering of the parameter space, and numerically effectively costless leapfrog steps. This shows how quick forecasts can be upgraded to accurate forecasts whenever needed. Results were gained with the public code from http://lnasellentin.github.io/DALI/, Comment: 9 pages, 7 figures

In this paper we extent the previously published DALI-approximation for likelihoods to cases in which the parameter dependency is in the covariance matrix. The approximation recovers non-Gaussian likelihoods, and reduces to the Fisher matrix approach in the case of Gaussianity. It works with the minimal assumptions of having Gaussian errors on the data, and a covariance matrix that possesses a converging Taylor approximation. The resulting approximation works in cases of severe parameter degeneracies and in cases where the Fisher matrix is singular. It is at least $1000$ times faster than a typical Monte Carlo Markov Chain run over the same parameter space. Two example applications, to cases of extremely non-Gaussian likelihoods, are presented -- one demonstrates how the method succeeds in reconstructing completely a ring-shaped likelihood. A public code is released here: http://lnasellentin.github.io/DALI/, Comment: accepted for publication in MNRAS

Three relativistic particles in addition to the photon are detected in the cosmic microwave background (CMB). In the standard model of cosmology, these are interpreted as the three neutrino species. However, at the time of CMB-decoupling, neutrinos are not only relativistic but they are also freestreaming. Here, we investigate, whether the CMB is sensitive to this defining feature of neutrinos, or whether the CMB-data allow to replace neutrinos with a relativistic fluid. We show that free streaming particles are preferred over a relativistic perfect fluid with $\Delta\chi^2\simeq 21$. We also study the possibility to replace the neutrinos by a viscous fluid and find that a relativistic viscous fluid with either the standard values $c_{\rm eff}^2=c_{\rm vis}^2=1/3$ or best fit values for $c_{\rm eff}^2$ and $c_{\rm vis}^2$ has $\Delta\chi^2=20$ and thus cannot provide a good fit to present CMB data either., Comment: 10 pages 7 figures, version accepted for publication in PRD

We present the new method DALI (Derivative Approximation for LIkelihoods) for reconstructing and forecasting posteriors. DALI extends the Fisher Matrix formalism but allows for a much wider range of posterior shapes. While the Fisher Matrix formalism is limited to yield ellipsoidal confidence contours, our method can reproduce the often observed flexed, deformed or curved shapes of known posteriors. This gain in shape fidelity is obtained by expanding the posterior to higher order in derivatives with respect to parameters, such that non-Gaussianity in the parameter space is taken into account. The resulting expansion is positive definite and normalizable at every order. Here, we present the new technique, highlight its advantages and limitations, and show a representative application to a posterior of dark energy parameters from supernovae measurements., Comment: v3: Corrected an error in Appendix A. Results and conclusions unchanged. v2: New figures and references added. Typos corrected. Matches published version. 10 pages, 6 figures