Partition function approach to non-Gaussian likelihoods: Formalism and expansions for weakly non-Gaussian cosmological inference

Abstract

Non-Gaussian likelihoods, ubiquitous throughout cosmology, are a direct consequence of nonlinearities in the physical model. Their treatment requires Monte-Carlo Markov-chain or more advanced sampling methods for the determination of confidence contours. As an alternative, we construct canonical partition functions as Laplace-transforms of the Bayesian evidence, from which MCMC-methods would sample microstates. Cumulants of order $n$ of the posterior distribution follow by direct $n$-fold differentiation of the logarithmic partition function, recovering the classic Fisher-matrix formalism at second order. We connect this approach for weakly non-Gaussianities to the DALI- and Gram-Charlier expansions and demonstrate the validity with a supernova-likelihood on the cosmological parameters ${\Omega }_m$ and $w$. We comment on extensions of the canonical partition function to include kinetic energies in order to bridge to Hamilton Monte-Carlo sampling, and on ensemble Markov-chain methods, as they would result from transitioning to macrocanonical partition functions depending on a chemical potential. Lastly we demonstrate the relationship of the partition function approach to the Cramér-Rao boundary and to information entropies.

arXiv comment12 pages 2 figures, Affiliation updated

References in corpus (45)

• Observational Evidence from Supernovae for an Accelerating Universe and a Cosmological Constant
• emcee: The MCMC Hammer
• Cosmological parameters from CMB and other data: a Monte-Carlo approach
• The Hubble Space Telescope Cluster Supernova Survey: V. Improving the Dark Energy Constraints Above z>1 and Building an Early-Type-Hosted Supernova Sample
• Rényi Divergence and Kullback-Leibler Divergence
• Improved Cosmological Constraints from New, Old and Combined Supernova Datasets
• Spectra and Light Curves of Six Type Ia Supernovae at 0.511 < z < 1.12 and the Union2 Compilation
• Bayes in the sky: Bayesian inference and model selection in cosmology
• Karhunen-Loeve eigenvalue problems in cosmology: how should we tackle large data sets?
• Quintessence: A Review
• Applications of Bayesian model selection to cosmological parameters
• CFHTLenS revisited: assessing concordance with Planck including astrophysical systematics
• Concordance and Discordance in Cosmology
• Systematic Bias in Cosmic Shear: Beyond the Fisher Matrix
• Bayesian reconstruction of the cosmological large-scale structure: methodology, inverse algorithms and numerical optimization
• Weakly Non-Linear Gaussian Fluctuations and the Edgeworth Expansion
• Fast Hamiltonian sampling for large scale structure inference
• Supernova cosmology: legacy and future
• Present and future evidence for evolving dark energy
• Bayesian Evidence and Model Selection
• Constraints on deviations from $\Lambda$CDM within Horndeski gravity
• Quantifying dimensionality: Bayesian cosmological model complexities
• On the Validity of Cosmological Fisher Matrix Forecasts
• Breaking the spell of Gaussianity: forecasting with higher order Fisher matrices
• iCosmo: an Interactive Cosmology Package
• Quantifying concordance of correlated cosmological data sets
• The power of Bayesian evidence in astronomy
• Magnification-Temperature Correlation: the Dark Side of ISW Measurements
• Cosmological Systematics Beyond Nuisance Parameters : Form Filling Functions
• Consistency tests in cosmology using relative entropy
• Biases on the cosmological parameters and thermal SZ residuals
• Information Gains from Cosmological Probes
• A fast, always positive definite and normalizable approximation of non-Gaussian likelihoods
• Fisher Matrix Preloaded -- Fisher4Cast
• On Model Selection in Cosmology
• Weak lensing tomography with orthogonal polynomials
• Quintessence and the Underlying Particle Physics Theory
• Path Integral Marginalization for Cosmology: Scale Dependent Galaxy Bias & Intrinsic Alignments
• Variations of cosmic large-scale structure covariance matrices across parameter space
• Describing variations of the Fisher-matrix across parameter space
• Figures of Merit for Testing Standard Models: Application to Dark Energy Experiments in Cosmology
• Fast calculation of the Fisher matrix for cosmic microwave background experiments
• Information geometry in cosmological inference problems
• Probe combination in large galaxy surveys : Application of Fisher information and Shannon entropy to weak lensing
• A test of the Suyama-Yamaguchi inequality from weak lensing

Cited by in corpus (1)

• Partition function approach to non-Gaussian likelihoods: macrocanonical partitions and replicating Markov-chains