Abstract:Gaussian approximations are routinely employed in Bayesian statistics to ease inference when the target posterior is intractable. Although these approximations are asymptotically justified by Bernstein-von Mises type results, in practice the expected Gaussian behavior may poorly represent the shape of the posterior, thus affecting approximation accuracy. Motivated by these considerations, we derive an improved class of closed-form approximations of posterior distributions which arise from a new treatment of a third-order version of the Laplace method yielding approximations in a tractable family of skew-symmetric distributions. Under general assumptions which account for misspecified models and non-i.i.d. settings, this family of approximations is shown to have a total variation distance from the target posterior whose rate of convergence improves by at least one order of magnitude the one established by the classical Bernstein-von Mises theorem. Specializing this result to the case of regular parametric models shows that the same improvement in approximation accuracy can be also derived for polynomially bounded posterior functionals. Unlike other higher-order approximations, our results prove that it is possible to derive closed-form and valid densities which are expected to provide, in practice, a more accurate, yet similarly-tractable, alternative to Gaussian approximations of the target posterior, while inheriting its limiting frequentist properties. We strengthen such arguments by developing a practical skew-modal approximation for both joint and marginal posteriors that achieves the same theoretical guarantees of its theoretical counterpart by replacing the unknown model parameters with the corresponding MAP estimate. Empirical studies confirm that our theoretical results closely match the remarkable performance observed in practice, even in finite, possibly small, sample regimes.

Approximating the mode of the non-central chi-squared distribution

Refined normal approximations for the central and noncentral chi-square distributions and some applications

A branch cut approach to the probability density and distribution functions of a linear combination of central and non-central Chi-square random variables

Novel PDF Approximations and Their Applications in Wireless Signal Detection

On the product of correlated normal random variables and the noncentral chi-square difference distribution

New methods to compute the generalized chi-square distribution

Non-asymptotic approximations for Pearson's chi-square statistic and its application to confidence intervals for strictly convex functions of the probability weights of discrete distributions

A note on the convergence rate of the kernel density estimator of the mode

On the computation of moments of the partial non-central chi-squared distribution function

Mixture representations of noncentral distributions

Approximate Distribution of Log 2 ( A + Χ 2 ) and Its Applications

Chi-Square Approximation for the Distribution of Individual Eigenvalues of a Singular Wishart Matrix

$\Chi^2$ Random Fields in Space and Time

A New Maximum Entropy Method for Estimation of Multimodal Probability Density Function

New asymptotic representations of the noncentral $t$-distribution

The estimate of $χ^2$ distance between binomial and generalized binomial distributions

The nonparametric Fisher geometry and the chi-square process density prior

Number Density of Peaks in a Chi-Squared Field

A study on the Weibull and Pareto distributions motivated by Chvátal's theorem

Skewed Bernstein-von Mises theorem and skew-modal approximations