Exploiting low-rank covariance structures for computing high-dimensional normal and Student-t probabilities
From MaRDI portal
Publication:2029072
Abstract: We present a preconditioned Monte Carlo method for computing high-dimensional multivariate normal and Student- probabilities arising in spatial statistics. The approach combines a tile-low-rank representation of covariance matrices with a block-reordering scheme for efficient Quasi-Monte Carlo simulation. The tile-low-rank representation decomposes the high-dimensional problem into many diagonal-block-size problems and low-rank connections. The block-reordering scheme reorders between and within the diagonal blocks to reduce the impact of integration variables from right to left, thus improving the Monte Carlo convergence rate. Simulations up to dimension suggest that the new method can improve the run time by an order of magnitude compared with the non-reordered tile-low-rank Quasi-Monte Carlo method and two orders of magnitude compared with the dense Quasi-Monte Carlo method. Our method also forms a strong substitute for the approximate conditioning methods as a more robust estimation with error guarantees. An application study to wind stochastic generators is provided to illustrate that the new computational method makes the maximum likelihood estimation feasible for high-dimensional skew-normal random fields.
Recommendations
- Hierarchical-block conditioning approximations for high-dimensional multivariate normal probabilities
- Fast computation of high-dimensional multivariate normal probabilities
- scientific article; zbMATH DE number 1304526
- Numerical computation of multivariate normal probabilities using bivariate conditioning
- Monte Carlo evaluation of multivariate Student's t probabilities
Cites work
- scientific article; zbMATH DE number 2174552 (Why is no real title available?)
- A unified view on skewed distributions arising from selections
- Algorithm AS 195: Multivariate Normal Probabilities with Error Bound
- Bivariate conditioning approximations for multivariate normal probabilities
- Computation of multivariate normal and \(t\) probabilities
- Conjugate Bayes for probit regression via unified skew-normal distributions
- Definition and probabilistic properties of skew-distributions.
- Estimating Orthant Probabilities of High-Dimensional Gaussian Vectors with An Application to Set Estimation
- Excursion and contour uncertainty regions for latent Gaussian models
- Hierarchical Decompositions for the Computation of High-Dimensional Multivariate Normal Probabilities
- Hierarchical matrices: algorithms and analysis
- Hierarchical matrix operations on GPUs. Matrix-vector multiplication and compression
- Hierarchical-block conditioning approximations for high-dimensional multivariate normal probabilities
- Introduction to hierarchical matrices with applications.
- Multivariate unified skew-elliptical distributions
- Numerical computation of multivariatet-probabilities with application to power calculation of multiple contrasts
- Parallel \(\mathcal H\)-matrix arithmetics on shared memory systems
- Parallel black box $$\mathcal {H}$$ -LU preconditioning for elliptic boundary value problems
- Principles for statistical inference on big spatio-temporal data from climate models
- Reducing storage of global wind ensembles with stochastic generators
- Some variants of the controlled random search algorithm for global optimization
- Statistical Applications of the Multivariate Skew Normal Distribution
- The no-U-turn sampler: adaptively setting path lengths in Hamiltonian Monte Carlo
- The normal law under linear restrictions: simulation and estimation via minimax tilting
- Updating Quasi-Newton Matrices with Limited Storage
Cited in
(5)- Tractable Bayes of Skew-Elliptical Link Models for Correlated Binary Data
- An EM algorithm for estimating the parameters of the multivariate skew-normal distribution with censored responses
- scientific article; zbMATH DE number 7625183 (Why is no real title available?)
- Bayesian Conjugacy in Probit, Tobit, Multinomial Probit and Extensions: A Review and New Results
- Scalable Computation of Predictive Probabilities in Probit Models with Gaussian Process Priors
This page was built for publication: Exploiting low-rank covariance structures for computing high-dimensional normal and Student-\(t\) probabilities
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2029072)
