Abstract
Practical Bayesian statistics with realistic models usually gives posterior distributions that are analytically intractable, and inferences must be made via numerical integration. In many cases, the integrands can be transformed into periodic functions on the unit $d$-dimensional cube, for which quasirandom sequences are known to give efficient numerical integration rules. This paper reviews some relevant theory, defines new criteria for identifying suitable quasirandom sequences and suggests some extensions to the basic integration rules. Various quasirandom methods are then compared on the sort of integrals that arise in Bayesian inference and are shown to be much more efficient than Monte Carlo methods.
Keywords
MathematicsUnit cubeNumerical integrationBayesian probabilitysortBayesian inferenceInferenceApplied mathematicsBayesian statisticsStatistical inferenceAlgorithmComputer scienceStatisticsDiscrete mathematicsArtificial intelligenceMathematical analysis
Related Publications
Three‐dimensional numerical integration for electronic structure calculations
Abstract Two three‐dimensional numerical schemes are presented for molecular integrands such as matrix alements of one‐electron operators occuring in the Fock operator and expec...
On the Numerical Calculation of the Density of States and Related Properties
Abstract A method is given for the numerical calculation of energy surface integrals within the Brillouin zone like density of states, conductivity, susceptibility, dielectric f...
Applications of a Method for the Efficient Computation of Posterior Distributions
For routine implementation with complicated likelihood functions, statistical procedures based on posterior distributions, or integrated likelihoods, require an efficient approa...
Progress with Numerical and Graphical Methods for Practical Bayesian Statistics
One of the main obstacles to the routine implementation of Bayesian methods has been the absence of efficient algorithms for carrying out the computational tasks implicit in the...
<i>Stan</i>: A Probabilistic Programming Language
Stan is a probabilistic programming language for specifying statistical models. A Stan program imperatively defines a log probability function over parameters conditioned on spe...
Publication Info
- Year
- 1988
- Type
- article
- Volume
- 16
- Issue
- 2
- Citations
- 98
- Access
- Closed
External Links
Social Impact
Altmetric
PlumX Metrics
Social media, news, blog, policy document mentions
Citation Metrics
98
OpenAlex
Cite This
J.E. Shaw
(1988).
A Quasirandom Approach to Integration in Bayesian Statistics.
The Annals of Statistics
, 16
(2)
.
https://doi.org/10.1214/aos/1176350842
Identifiers
- DOI
- 10.1214/aos/1176350842