This paper considers the evaluation of probabilities defined in terms of the multivariate normal distribution. The multivariate normal distribution can have any covariance matrix and any mean vector. Probabilities defined by a set of inequalities of linear combinations of the multivariate normal random variables are considered. It is shown how these probabilities can be evaluated from a series of one-dimensional integrations. This approach affords a practical algorithm for the evaluation of these probabilities which is considerably more efficient than more direct numerical integration approaches. Consequently, it enlarges the class of probabilities of this kind which are computationally feasible.
AMS Subject Classification
Multivariate normal distribution Recursive integration Numerical integration Computational intensity Tree structure Tri-diagonal matrix Orthant probability
This is a preview of subscription content, log in to check access.
Armitage, P., McPherson, C.K., Rowe, B.C., 1969. Repeated significance tests on accumulating data. Journal of the Royal Statistical Society, Series A, 132, 235–244.MathSciNetCrossRefGoogle Scholar
Bretz, F., Genz, A., Hothorn, L.A., 2001. On the numerical availability of multiple comparison procedures. Biometrical Journal, 43, 645–656.MathSciNetCrossRefGoogle Scholar
Genz, A., Bretz, F., 1999. Numerical computation of multivariate t probabilities with application to power calculation of multiple contrasts. Journal of Statistical Computation and Simulation, 63, 361–378.MathSciNetCrossRefGoogle Scholar
Genz, A., Bretz, F., 2002. Methods for the computation of multivariate t-probabilities. Journal of Computational and Graphical Statistics, 11, 950–971.MathSciNetCrossRefGoogle Scholar
Genz, A., Bretz, F., 2009. Computation of Multivariate Normal and t Probabilities. Springer.CrossRefGoogle Scholar
Hayter, A.J., 2006. Recursive integration methodologies with statistical applications. Journal of Statistical Planning and Inference, 136, 2284–2296.MathSciNetCrossRefGoogle Scholar
Hayter, A.J., Liu, W., 1996a. Exact calculations for the one-sided studentized range test for testing against a simple ordered alternative. Computational Statistics and Data Analysis, 22, 17–25.MathSciNetCrossRefGoogle Scholar
Hayter, A.J., Liu, W., 1996b. A note on the calculation of P(X1X2Xk). The American Statistician, 50(4), 365.Google Scholar
Hayter, A.J., Lin, Y., 2011. The evaluation of two-sided orthant probabilities for a quadrivariate normal distribution. To appear in Computational Statistics, 2012.MathSciNetCrossRefGoogle Scholar
Miwa, T., Hayter, A.J., Kuriki, S., 2003. The evaluation of general non-centred orthant probabilities. Journal of the Royal Statistical Society, Series B, 65, 223–234.MathSciNetCrossRefGoogle Scholar
Somerville, P.N., 1997. Multiple testing and simultaneous confidence intervals: calculation of constants. Computational Statistics and Data Analysis, 25, 217–233.CrossRefGoogle Scholar
Somerville, P.N., 1998. Numerical computation of multivariate normal and multivariate t probabilities over convex regions. J. Computational and Graphical Statistics, 7, 529–544.MathSciNetGoogle Scholar