A novel series expansion for the multivariate normal probability integrals based on Fourier series

Fayed, Hatem; Atiya, Amir

doi:10.1090/S0025-5718-2014-02844-5

A novel series expansion for the multivariate normal probability integrals based on Fourier series
HTML articles powered by AMS MathViewer

by Hatem A. Fayed and Amir F. Atiya PDF

Math. Comp. 83 (2014), 2385-2402 Request permission

Abstract:

In this article, we derive a series expansion of the multivariate normal probability integrals based on Fourier series. The basic idea is to transform the limits of each integral from $h_i$ to $\infty$ to be from $-\infty$ to $\infty$ by multiplying the integrand by a periodic square wave that approximates the domain of the integral. This square wave is expressed by its Fourier series expansion. Then a Cholesky decomposition of the covariance matrix is applied to transform the integrand to a simple one that can be easily evaluated. The resultant formula has a simple pattern that is expressed as multiple series expansion of trigonometric and exponential functions.

References

M. Abramowitz, I.A. Stegun, Handbook of Mathematical Functions, Dover, New York, 1964.
I. Deák, Three digit accurate multiple normal probabilities, Numer. Math. 35 (1980), no. 4, 369–380. MR 593834, DOI 10.1007/BF01399006
István Deák, Random number generators and simulation, Mathematical Methods of Operations Research, vol. 4, Akadémiai Kiadó (Publishing House of the Hungarian Academy of Sciences), Budapest, 1990. Translated and revised from the Hungarian by the author. MR 1080965
D. R. Divgi, Calculation of univariate and bivariate normal probability functions, Ann. Statist. 7 (1979), no. 4, 903–910. MR 532253
T.G. Donnelly, Algorithm 462: Bivariate normal distribution, Commun. ACM 16 (1973) 636.
Z. Drezner, Computation of the bivariate normal integral, Math. Comp. 32 (1978), no. 141, 277–279. MR 461849, DOI 10.1090/S0025-5718-1978-0461849-9
Zvi Drezner and G. O. Wesolowsky, On the computation of the bivariate normal integral, J. Statist. Comput. Simulation 35 (1990), no. 1-2, 101–107. MR 1041725, DOI 10.1080/00949659008811236
Hatem A. Fayed and Amir F. Atiya, An evaluation of the integral of the product of the error function and the normal probability density with application to the bivariate normal integral, Math. Comp. 83 (2014), no. 285, 235–250. MR 3120588, DOI 10.1090/S0025-5718-2013-02720-2
J. Gai, A computational study of the bivariate normal probability function, M.Sc. thesis, Department of Mathematics and Statistics, Queen’s University, Kingston, Ontario, Canada, 2002.
H. I. Gassmann, Multivariate normal probabilities: implementing an old idea of Plackett’s, J. Comput. Graph. Statist. 12 (2003), no. 3, 731–752. MR 2005459, DOI 10.1198/1061860032283
A. Genz, Numerical computation of multivariate normal probabilities, J. Comp. Graph. Stat. 1 (1992) 141-150.
A. Genz, Comparison of methods for the computation of multivariate normal probabilities, Comp. Sci. Stat. 25 (1993) 400-405.
Alan Genz, Numerical computation of rectangular bivariate and trivariate normal and $t$ probabilities, Stat. Comput. 14 (2004), no. 3, 251–260. MR 2086401, DOI 10.1023/B:STCO.0000035304.20635.31
Alan Genz and Frank Bretz, Computation of multivariate normal and $t$ probabilities, Lecture Notes in Statistics, vol. 195, Springer, Dordrecht, 2009. MR 2840595, DOI 10.1007/978-3-642-01689-9
Shanti S. Gupta, Probability integrals of multivariate normal and multivariate $t$, Ann. Math. Statist. 34 (1963), 792–828. MR 152068, DOI 10.1214/aoms/1177704004
Bernard Harris and Andrew P. Soms, The use of the tetrachoric series for evaluating multivariate normal probabilities, J. Multivariate Anal. 10 (1980), no. 2, 252–267. MR 575928, DOI 10.1016/0047-259X(80)90017-2
N.L. Johnson, S. Kotz, Distributions in statistics: Continuous multivariate distribution, John Wiley and Sons, New York, 1972.
M. G. Kendall, Proof of relations connected with the tetrachoric series and its generalization, Biometrika 32 (1941), 196–198. MR 5573, DOI 10.1093/biomet/32.2.196
Tetsuhisa Miwa, A. J. Hayter, and Satoshi Kuriki, The evaluation of general non-centred orthant probabilities, J. R. Stat. Soc. Ser. B Stat. Methodol. 65 (2003), no. 1, 223–234. MR 1959823, DOI 10.1111/1467-9868.00382
Donald B. Owen, Tables for computing bivariate normal probabilities, Ann. Math. Statist. 27 (1956), 1075–1090. MR 127562, DOI 10.1214/aoms/1177728074
K. Pearson, Mathematical contributions to the theory of evolution. VII. on the correlation of characters not quantitatively. Philos. Trans. R. Soc. S-A. 196 (1901) 1-47.
A. P. Prudnikov, Yu. A. Brychkov, and O. I. Marichev, Integrals and series. Vol. 1, Gordon & Breach Science Publishers, New York, 1986. Elementary functions; Translated from the Russian and with a preface by N. M. Queen. MR 874986
N. G. Shephard, From characteristic function to distribution function: a simple framework for the theory, Econometric Theory 7 (1991), no. 4, 519–529. MR 1151944, DOI 10.1017/S0266466600004746
Tamás Szántai, An algorithm for determining the values of multivariate normal distributions and their gradients, Alkalmaz. Mat. Lapok 2 (1976), no. 1-2, 27–39 (Hungarian, with English summary). MR 438537
Y. L. Tong, The multivariate normal distribution, Springer Series in Statistics, Springer-Verlag, New York, 1990. MR 1029032, DOI 10.1007/978-1-4613-9655-0