Abstract
We compute upper and lower bounds on the expected maximum of correlated normal variables (up to a few hundred in number) with arbitrary means, variances, and correlations. Two types of bounding processes are used: perfectly dependent normal variables, and independent normal variables, both with arbitrary mean values. The expected maximum for the perfectly dependent variables can be evaluated in closed form; for the independent variables, a single numerical integration is required. Higher moments are also available. We use mathematical programming to find parameters for the processes, so they will give bounds on the expected maximum, rather than approximations of unknown accuracy. Our original application is to the maximum number of people on-line simultaneously during the day in an infinite-server queue with a time-varying arrival rate. The upper and lower bounds are tighter than previous bounds, and in many of our examples are within 5% or 10% of each other. We also demonstrate the bounds’ performance on some PERT models, AR/MA time series, Brownian motion, and product-form correlation matrices.
Similar content being viewed by others
References
Adler RJ, Taylor JE (2007) Random fields and their geometry. Springer
Azaïs J-M, Wschebor M (2001) Computing the distribution of the maximum of a gaussian process. Technical Report 01/44, Mathematics Dept., Universidad de la Republica, Uruguay
Berman SM (1992) Sojourns and extremes of stochastic processes. Wadsworth and Brooks/Cole
Bose RC, Gupta SS (1959) Moments of order statistics from a normal population. Biometrika 46(3/4):433–440, Dec
Brockwell PJ, Davis RA (1996) Introduction to time series and forecasting. Springer Texts in Statistics
Brodtkorb PA (2006) Evaluating nearly singular multinormal expectations with application to wave distributions. Methodol Comput Appl Probab 8(1):65–91
Brown LD, Gans N, Mandelbaum A, Sakov A, Shen H, Zeltyn S, Zhao LH (2005) Statistical analysis of a telephone call center: a queueing-science perspective. J Am Stat Assoc 100(469):36–50, Mar
Chinneck JW (1998) Analyzing mathematical programs using MProbe. Technical Report SCE-98-03, Systems and Computer Engineering, Carleton Univ. Software available at http://www.sce.carleton.ca/faculty/chinneck/mprobe.html
Cierco-Ayrolles C, Croquette A, Delmas C (2003) Computing the distribution of the maximum of gaussian random processes. Methodol Comput Appl Probab 5(4):427–438, Dec
Clark CE (1961) The greatest of a finite set of random variables. Oper Res 9(2):145–162, Mar–Apr
Clark CE, Trevor Williams G (1958) Distributions of the members of an ordered sample. Ann Math Stat 29(3):862–870, Sep
Coles SG (2001) An introduction to statistical modeling of extreme values. Springer Series in Statistics. Springer
Cox DR (1955) Some statistical methods connected with series of events. J Royal Stat Soc B 17:129–164
David HA (1981) Order statistics, 2nd edn. Wiley
Eick SG, Massey WA, Whitt W (1993a) M t /G/ ∞ queues with sinusoidal arrival rates. Manage Sci 39(2):241–252, Feb
Eick SG, Massey WA, Whitt W (1993b) The physics of the M t /G/ ∞ queue. Oper Res 41(4):731–742, Jul–Aug
Genz A (1992) Numerical computation of multivariate normal probabilities. J Comput Graph Stat 1:141–149
Genz A, Monahan J (1999) A stochastic algorithm for high dimensional multiple integrals over unbounded regions with Gaussian weight. J Comput Appl Math 112(1–2):71–81, Nov
Grandell J (1997) Mixed poisson processes. Number 77 in monographs on statistics and applied probability. Chapman and Hall
Green LV, Kolesar PJ, Svoronos A (1991) Some effects of nonstationarity on multiserver Markovian queueing systems. Oper Res 39(3):502–511, May–Jun
Henderson SG, Chen BPK (2001) Two issues in setting call centre staffing levels. Ann Oper Res 108(1–4):176–192, Nov
Hillier FS, Lieberman GJ (1995) Introduction to operations research, 7th edn. McGraw-Hill
Kang S, Serfozo RF (1999) Extreme values of phase-type and mixed random variables with parallel-processing examples. J Appl Probab 36(1):194–210, Jun
Lai TL, Robbins H (1976) Maximally dependent random variables. Proceedings of the National Academy of the Sciences USA 73(2):286–288, Feb
Leadbetter MR, Lindgren G, Rootzen H (1983) Extremes and related properties of random sequences and processes. Springer Series in Statistics. Springer
Malcolm DG, Roseboom JH, Clark CE, Fazar W (1959) Application of a technique for research and development program evaluation. Oper Res 7:646–669
Owen DB, Steck GP (1962) Moments of order statistics from the equicorrelated multivariate normal distribution. Ann Math Stat 33(4):1286–1291, Dec
Palm C (1988) Intensity variations in telephone traffic (translation of 1943 article in Ericcson Technics, vol 44, pp 1–189). North-Holland, Amsterdam
Ross AM (2003a) Useful bounds on the expected maximum of correlated normal variables. Technical Report 03W-004, ISE Dept., Lehigh Univ., Aug
Ross SM (2003b) Introduction to probability models, 8th edn. Academic Press
Ross AM, George Shanthikumar J (2008) Dial-up internet access: a two-provider cost model. Queueing Syst 51:5–27
Slepian D (1962) The one-sided barrier problem for Gaussian processes. Bell Syst Tech J 41:463–501
Teichroew D (1956) Tables of expected values of order statistics and products of order statistics for samples of size twenty or less from the normal distribution. Ann Math Stat 27(2):410–426, Jun
Thompson GM (1999) Setting staffing levels in pure service environments when the true mean daily customer arrival rate is a normal random variate (working paper)
Tippett LHC (1925) On the extreme individuals and the range of samples taken from a normal population. Biometrika 17(3/4):364–387, Dec
Tong YL (1990) The multivariate normal distribution. Springer
Vitale RA (2000) Some comparisons for Gaussian processes. Proceedings of the American Math. Society 128(10):3043–3046
Yates R, Goodman D (2005) Probability and stochastic processes: a friendly introduction for electrical and computer engineers, 2nd edn. Wiley
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Ross, A.M. Computing Bounds on the Expected Maximum of Correlated Normal Variables. Methodol Comput Appl Probab 12, 111–138 (2010). https://doi.org/10.1007/s11009-008-9097-z
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11009-008-9097-z