Abstract
New formulas are derived for multivariate normal probabilities defined for hyper-rectangular probability regions. The formulas use conditioning with a sequence of bivariate normal probabilities. The result is an approximate formula for multivariate normal probabilities which uses a product of bivariate normal probabilities. The new approximation method is compared with approximation methods based on products of univariate normal probabilities, using tests with random covariance-matrix/probability-region problems for up to twenty variables. The reordering of variables is studied to improve efficiency of the new method.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Connors, R.D., Hess, S., Daly, A.: Analytic approximations for computing probit choice probabilities. Transportmetrica 10(2), 119–139 (2014)
Drezner, Z., Wesolowsky, G.O.: On the computation of the bivariate normal integral. J. Stat. Comput. Simul. 3, 101–107 (1990)
Genz, A.: Numerical computation of rectangular bivariate and trivariate normal and \(t\) probabilities. Stat. Comput. 14, 151–160 (2004)
Genz, A.: MVNXPB, a MATLAB/Octave function for the approximation of multivariate Normal probabilities. (2013), at www.math.wsu.edu/faculty/genz/software/matlab
Genz, A., Bretz, F.: Computation of Multivariate Normal and t Probabilities. Lecture Notes in Statistics 195, Springer, New York (2009)
Golub, G.H., Van Loan, C.F.: Matrix Computations, 4th edn. Johns Hopkins University Press, Baltimore (2013)
Gibson, G.J., Glasbey, C.A., Elston, D.A.: Monte Carlo evaluation of multivariate normal integrals and sensitivity to variate ordering. In: Dimov, I.T., Sendov, B., Vassilevski, P.S. (eds.) Advances in Numerical Methods and Applications, pp. 120–126. World Scientific Publishing, River Edge (1994)
Ho, H.J., Lin, T.I., Chen, H.Y., Wang, W.L.: Some results on the truncated multivariate t distribution. J. Stat. Plan. Inference 142, 25–40 (2012)
Joe, H.: Approximations to multivariate normal rectangle probabilities based on conditional expectations. J. Am. Stat. Assoc. 90, 957–964 (1995)
Kamakura, W.A.: The estimation of multinomial probit models: a new calibration algorithm. Transp. Sci. 23, 253–265 (1989)
McFadden, D., Train, K.: Mixed MNL models for discrete response. J. Appl. Econom. 15, 447–470 (2000)
Mendell, N.R., Elston, R.C.: Multifactorial qualitative traits: genetic analysis and prediction of recurrence risks. Biometrics 30, 41–57 (1974)
Muthén, B.: Moments of the censored and truncated bivariate normal distribution. Br J Math Stat Psychol 43, 131–143 (1991)
Ochi, Y., Prentice, R.L.: Likelihood inference in a correlated probit regression model. Biometrika 71, 531–543 (1984)
Rosenbaum, S.: Moments of a truncated bivariate normal distribution. J. R. Stat. Soc. 23, 405–408 (1961)
Schervish, M.J.: Algorithm AS 195: multivariate normal probabilities with error bound. J. R. Stat. Soc. Ser. C 33, 81–94 (1984), correction 34, 103–104 (1985)
Stewart, G.W.: The efficient generation of random orthogonal matrices with an application to condition estimators. SIAM J. Numer. Anal. 17(3), 403–409 (1980)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Trinh, G., Genz, A. Bivariate conditioning approximations for multivariate normal probabilities. Stat Comput 25, 989–996 (2015). https://doi.org/10.1007/s11222-014-9468-y
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11222-014-9468-y