×

zbMATH — the first resource for mathematics

Random sampling from a truncated multivariate normal distribution. (English) Zbl 0795.62060
Summary: This note critiques a procedure suggested by K. E. Ahmad and N. S. Abd-El-Hakim [ibid. 3, No. 3, 1-3 (1990; Zbl 0706.60014)] for drawing random samples from a truncated bivariate normal distribution. An alternative methodology using the Gibbs sampler is described. The technique can be used for sampling from a truncated multivariate normal distribution. The Gauss code is included as an appendix.

MSC:
62H99 Multivariate analysis
62E99 Statistical distribution theory
62-04 Software, source code, etc. for problems pertaining to statistics
PDF BibTeX XML Cite
Full Text: DOI
References:
[1] Ahmad, K.E.; Abd-El-Hakim, N.S., Random sampling from a truncated bivariate normal distribution, Appl. math. lett., 3, 3, 1-3, (1990) · Zbl 0706.60014
[2] Tadikamalla, P.R.; Johnson, M.E., Simple rejection methods for sampling from the normal distribution, Proceedings of the first international conference of mathematical modeling, 573-577, (1977)
[3] Hajivassiliou, V.A.; McFadden, D.L., The method of simulated scores for the estimation of LDV models with an application to external debt crises, Cowles foundation discussion paper no. 967, (1990), Yale University CT
[4] Geman, S.; Geman, D., Stochastic relaxation, Gibbs distributions and the Bayesian restoration of images, IEEE transactions on pattern analysis and machine intelligence, 6, 721-741, (1984) · Zbl 0573.62030
[5] Tanner, T.; Wong, W., The calculation of posterior distributions by data augmentation, Journal of the American statistical association, 82, 528-549, (1987) · Zbl 0619.62029
[6] Hajivassiliou, V.A.; McFadden, D.L.; Rudd, P., Simulation of multivariate normal orthant probabilities: methods and programs, Working paper, (1991), Yale University CT
[7] McCulloch, R.; Rossi, P.E., An exact likelihood analysis of the multinomial probit model, Working paper 91-102, (1992), Dept. of Economics, University of Chicago
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.