Access

You are not currently logged in.

Access your personal account or get JSTOR access through your library or other institution:

login

Log in to your personal account or through your institution.

Computing Multivariate Normal Probabilities: A New Look

H. I. Gassmann, I. Deák and T. Szántai
Journal of Computational and Graphical Statistics
Vol. 11, No. 4 (Dec., 2002), pp. 920-949
Stable URL: http://www.jstor.org/stable/1391170
Page Count: 30
  • Download ($14.00)
  • Cite this Item
Computing Multivariate Normal Probabilities: A New Look
Preview not available

Abstract

This article describes and compares several numerical methods for finding multivariate probabilities over a rectangle. A large computational study shows how the computation times depend on the problem dimensions, the correlation structure, the magnitude of the sought probability, and the required accuracy. No method is uniformly best for all problems and--unlike previous work--this article gives some guidelines to help establish the most promising method a priori. Numerical tests were conducted on approximately 3,000 problems generated randomly in up to 20 dimensions. Our findings indicate that direct integration methods give acceptable results for up to 12-dimensional problems, provided that the probability mass of the rectangle is not too large (less than about 0.9). For problems with small probabilities (less than 0.3) a crude Monte Carlo method gives reasonable results quickly, while bounding procedures perform best on problems with large probabilities (> 0.9). For larger problems numerical integration with quasirandom Korobov points may be considered, as may a decomposition method due to Deák. The best method found four-digit accurate probabilities for every 20-dimensional problem in less than six minutes on a 533MHz Pentium III computer.

Page Thumbnails

  • Thumbnail: Page 
920
    920
  • Thumbnail: Page 
921
    921
  • Thumbnail: Page 
922
    922
  • Thumbnail: Page 
923
    923
  • Thumbnail: Page 
924
    924
  • Thumbnail: Page 
925
    925
  • Thumbnail: Page 
926
    926
  • Thumbnail: Page 
927
    927
  • Thumbnail: Page 
928
    928
  • Thumbnail: Page 
929
    929
  • Thumbnail: Page 
930
    930
  • Thumbnail: Page 
931
    931
  • Thumbnail: Page 
932
    932
  • Thumbnail: Page 
933
    933
  • Thumbnail: Page 
934
    934
  • Thumbnail: Page 
935
    935
  • Thumbnail: Page 
936
    936
  • Thumbnail: Page 
937
    937
  • Thumbnail: Page 
938
    938
  • Thumbnail: Page 
939
    939
  • Thumbnail: Page 
940
    940
  • Thumbnail: Page 
941
    941
  • Thumbnail: Page 
942
    942
  • Thumbnail: Page 
943
    943
  • Thumbnail: Page 
944
    944
  • Thumbnail: Page 
945
    945
  • Thumbnail: Page 
946
    946
  • Thumbnail: Page 
947
    947
  • Thumbnail: Page 
948
    948
  • Thumbnail: Page 
949
    949