The normal law under linear restrictions: simulation and estimation via minimax tilting

@article{Botev2016TheNL,
  title={The normal law under linear restrictions: simulation and estimation via minimax tilting},
  author={Zdravko I. Botev},
  journal={Journal of the Royal Statistical Society: Series B (Statistical Methodology)},
  year={2016},
  volume={79}
}
  • Z. Botev
  • Published 14 March 2016
  • Mathematics
  • Journal of the Royal Statistical Society: Series B (Statistical Methodology)
Simulation from the truncated multivariate normal distribution in high dimensions is a recurrent problem in statistical computing and is typically only feasible by using approximate Markov chain Monte Carlo sampling. We propose a minimax tilting method for exact independently and identically distributed data simulation from the truncated multivariate normal distribution. The new methodology provides both a method for simulation and an efficient estimator to hitherto intractable Gaussian… 

LOGARITHMICALLY EFFICIENT ESTIMATION OF THE TAIL OF THE MULTIVARIATE NORMAL DISTRIBUTION

Simulation from the tail of the multivariate normal density has numerous applications in statistics and operations research. Unfortunately, there is no simple formula for the cumulative distribution

Logarithmically efficient estimation of the tail of the multivariate normal distribution

An asymptotically efficient Monte Carlo estimator for quantities related to the tail of the multivariate normal distribution and a simple and practical Markov chain sampler for the normal tail that is asymPTotically optimal is proposed.

Efficient estimation and simulation of the truncated multivariate student-\(t\) distribution

We propose an exponential tilting method for exact simulation from the truncated multivariate student-t distribution in high dimensions as an alternative to approximate Markov Chain Monte Carlo

Finite-dimensional Gaussian approximation with linear inequality constraints

The finite-dimensional Gaussian approach from Maatouk and Bay (2017) is considered which can satisfy inequality conditions everywhere and its full framework together with a Hamiltonian Monte Carlo-based sampler provides efficient results on both data fitting and uncertainty quantification.

Efficient probability estimation and simulation of the truncated multivariate student-t distribution

An exponential tilting method for the accurate estimation of the probability that a random vector with multivariate student-t distribution falls in a convex polytope is proposed, providing an alternative to approximate Markov Chain Monte Carlo simulation.

EXACT POSTERIOR SIMULATION FROM THE LINEAR LASSO REGRESSION

This paper constructs a method that yields an independent and identically distributed (iid) draws from the LASSO posterior that works well when the dimension of the parameter space is not too large, and when it is too large to permit exact sampling, the proposed method can still be used to construct an approximate MCMC sampler.

EXACT POSTERIOR SIMULATION FROM THE LINEAR LASSO REGRESSION

This paper constructs a method that yields an independent and identically distributed (iid) draws from the LASSO posterior that works well when the dimension of the parameter space is not too large, and when it is too large to permit exact sampling, the proposed method can still be used to construct an approximate MCMC sampler.

Fast and Exact Simulation of Multivariate Normal and Wishart Random Variables with Box Constraints

This work introduces computationally efficient methods to make exact and independent draws from both the multivariate normal and Wishart distributions with box constraints, and improves the feasibility of Monte Carlo-based inference for box-constrained, multivariatenormal and wishart distributions.

Simulation from the Normal Distribution Truncated to an Interval in the Tail

We study and compare various methods to generate a random variate from the normal distribution truncated to some finite or semi-infinite interval, with special attention to the situation where the

The Soft Multivariate Truncated Normal Distribution with Applications to Bayesian Constrained Estimation

The soft tMVN distribution can be used to approximate simulations from a multivariate truncated normal distribution with linear constraints, or itself as a prior in shape-constrained problems.
...

References

SHOWING 1-10 OF 43 REFERENCES

Efficient estimation and simulation of the truncated multivariate student-\(t\) distribution

We propose an exponential tilting method for exact simulation from the truncated multivariate student-t distribution in high dimensions as an alternative to approximate Markov Chain Monte Carlo

Fast simulation of truncated Gaussian distributions

  • N. Chopin
  • Computer Science, Mathematics
    Stat. Comput.
  • 2011
This work designs a table-based algorithm that is computationally faster than alternative algorithms and an accept-reject algorithm for simulating a Gaussian vector X, conditional on the fact that each component of X belongs to a finite interval, or a semi-finite interval.

Efficient inference for spatial extreme value processes associated to log-Gaussian random functions

This work considers random fields that are in the domain of attraction of a widely used class of max-stable processes, namely those constructed via manipulation of log-Gaussian random functions, and performs full likelihood inference by exploiting the methods of Stephenson & Tawn (2005), assessing the improvements in inference from both methods over pairwise likelihood methodology.

An accept-reject algorithm for the positive multivariate normal distribution

An accept-reject algorithm is introduced in which variates from a positive multivariate normal distribution are proposed from a multivariate skew-normal distribution and is shown, under certain conditions, to be very efficient.

Asymptotic robustness of estimators in rare-event simulation

The asymptotic robustness of estimators as a function of a rarity parameter, in the context of rare-event simulation, is often qualified by properties such as bounded relative error and logarithmic efficiency, which are developed in a setting where state-dependent importance sampling (IS) is used to estimate first-passage time probabilities.

Bounded normal approximation in simulations of highly reliable Markovian systems

  • B. Tuffin
  • Mathematics
    Journal of Applied Probability
  • 1999
In this paper, we give necessary and sufficient conditions to ensure the validity of confidence intervals, based on the central limit theorem, in simulations of highly reliable Markovian systems. We

Monte Carlo evaluation of multivariate normal probabilities

The method of simulated scores for the estimation of LDV models

The method of simulated scores (MSS) is presented for estimating limited dependent variables models (LDV) with flexible correlation structure in the unobservables. We propose simulators that are

Perfect simulation of positive Gaussian distributions

An exact simulation algorithm that produces variables from truncated Gaussian distributions on R via a perfect sampling scheme, based on stochastic ordering and slice sampling, since accept-reject algorithms like the one of Geweke and Robert are difficult to extend to higher dimensions.