A comparison of bivariate normal algorithms
Authors:
- 1 May 1991
- journal article
- research article
- Published by Taylor & Francis in Journal of Statistical Computation and Simulation
- Vol. 39 (1) , 115-127
Abstract
Statistical methods that implement numerical approximations to the Bivariate Normal (BN) distribution function require high accuracy. Most notable are those involving optimization (e.g., maximum likelihood and nonlinear least squares). Unfortunately high accuracy comes at the cost of increased computation time. For this reason, in the present paper we compare eight Bivariate Normal (BN) approximation algorithms with regard to the accuracy speed trade-off. The method developed by Divgi (1979) emerges as the clear method of choice, achieving 14-digit accuracy ten and a half times faster than its nearest competitor. Furthermore, in the time required by Divgi's approximation to reach this level of precision none of the other methods can support better than 3-digit accuracy.Keywords
This publication has 11 references indexed in Scilit:
- A Simple Approximation for Bivariate Normal ProbabilitiesJournal of Quality Technology, 1983
- Calculation of Univariate and Bivariate Normal Probability FunctionsThe Annals of Statistics, 1979
- A Polynomial Type Approximation for Bivariate Normal VariatesSIAM Journal on Applied Mathematics, 1978
- Computation of the bivariate normal integralMathematics of Computation, 1978
- Algorithm AS 76: An Integral Useful in Calculating Non-Central t and Bivariate Normal ProbabilitiesJournal of the Royal Statistical Society Series C: Applied Statistics, 1974
- Computation of Bi- and Tri-Variate Normal IntegralsJournal of the Royal Statistical Society Series C: Applied Statistics, 1974
- Algorithm 462: bivariate normal distributionCommunications of the ACM, 1973
- A Modification of Owen's Method for Computing the Bi-Variate Normal IntegralJournal of the Royal Statistical Society Series C: Applied Statistics, 1973
- Probability Content of Regions Under Spherical Normal Distributions, III: The Bivariate Normal IntegralThe Annals of Mathematical Statistics, 1961
- Tables for Computing Bivariate Normal ProbabilitiesThe Annals of Mathematical Statistics, 1956