• Corpus ID: 246822695

Adaptive truncation of infinite sums: applications to Statistics

@inproceedings{Carvalho2022AdaptiveTO,
  title={Adaptive truncation of infinite sums: applications to Statistics},
  author={Luiz Max Carvalho and Guido A. Moreira},
  year={2022}
}
It is often the case in Statistics that one needs to compute sums of infinite series, especially in marginalising over discrete latent variables. This has become more relevant with the popularization of gradient-based techniques (e.g. Hamiltonian Monte Carlo) in the Bayesian inference context, for which discrete latent variables are hard or impossible to deal with. For many major infinite series, like the Hurwitz Zeta function or the Conway-Maxwell Poisson normalising constant, custom algorithms… 

Figures and Tables from this paper

References

SHOWING 1-10 OF 46 REFERENCES

Rank-Normalization, Folding, and Localization: An Improved Rˆ for Assessing Convergence of MCMC (with Discussion)

A collection of quantile-based local efficiency measures, along with a practical approach for computing Monte Carlo error estimates for quantiles, are introduced and it is suggested that common trace plots should be replaced with rank plots from multiple chains.

A Function Emulation Approach for Doubly Intractable Distributions

A novel algorithm is proposed that provides computational gains over existing methods by replacing Monte Carlo approximations to the normalizing function with a Gaussian process-based approximation and is applicable more broadly to any likelihood function that is expensive to evaluate.

Markov Chain Truncation for Doubly-Intractable Inference

This work demonstrates how to construct unbiased estimates for 1/Z given access to black-box importance sampling estimators for Z, and adapts recent work on random series truncation and Markov chain coupling, producing estimators with lower variance and a higher percentage of positive estimates than before.

Double Exponential Families and Their Use in Generalized Linear Regression

Abstract In one-parameter exponential families such as the binomial and Poisson, the variance is a function of the mean. Double exponential families allow the introduction of a second parameter that

An adaptive truncation method for inference in Bayesian nonparametric models

This paper describes an adaptive truncation method which allows the level of the truncation to be decided by the algorithm and so can avoid large errors in approximating the posterior.

Stan: A Probabilistic Programming Language

Stan is a probabilistic programming language for specifying statistical models that provides full Bayesian inference for continuous-variable models through Markov chain Monte Carlo methods such as the No-U-Turn sampler and an adaptive form of Hamiltonian Monte Carlo sampling.

An asymptotic expansion for the normalizing constant of the Conway–Maxwell–Poisson distribution

The Conway–Maxwell–Poisson distribution is a two-parameter generalization of the Poisson distribution that can be used to model data that are under- or over-dispersed relative to the Poisson

Rater: An R Package for Fitting Statistical Models of Repeated Categorical Ratings

An R package, rater, available on CRAN, is presented that implements Bayesian versions of several statistical models that allow analysis of repeated categorical rating data and the techniques of marginalisation and conditioning are discussed.

Series evaluation of Tweedie exponential dispersion model densities

This article describes how the series expansions can be summed in an numerically efficient fashion and demonstrates the usefulness of the approach, but full machine accuracy is shown not to be obtainable using the series expansion method for all parameter values.