Estimating the parameters of a nonhomogeneous Poisson process with linear rate
 William A. Massey,
 Geraldine A. Parker,
 Ward Whitt
 … show all 3 hide
Rent the article at a discount
Rent now* Final gross prices may vary according to local VAT.
Get AccessAbstract
Motivated by telecommunication applications, we investigate ways to estimate the parameters of a nonhomogeneous Poisson process with linear rate over a finite interval, based on the number of counts in measurement subintervals. Such a linear arrivalrate function can serve as a component of a piecewiselinear approximation to a general arrivalrate function. We consider ordinary least squares (OLS), iterative weighted least squares (IWLS) and maximum likelihood (ML), all constrained to yield a nonnegative rate function. We prove that ML coincides with IWLS. As a reference point, we also consider the theoretically optimal weighted least squares (TWLS), which is least squares with weights inversely proportional to the variances (which would not be known with data). Overall, ML performs almost as well as TWLS. We describe computer simulations conducted to evaluate these estimation procedures. None of the procedures differ greatly when the rate function is not near 0 at either end, but when the rate function is near 0 at one end, TWLS and ML are significantly more effective than OLS. The number of measurement subintervals (with fixed total interval) makes surprisingly little difference when the rate function is not near 0 at either end. The variances are higher with only two or three subintervals, but there usually is little benefit from going above ten. In contrast, more measurement intervals help TWLS and ML when the rate function is near 0 at one end. We derive explicit formulas for the OLS variances and the asymptotic TWLS variances (as the number of measurement intervals increases), assuming the nonnegativity constraints are not violated. These formulas reveal the statistical precision of the estimators and the influence of the parameters and the method. Knowing how the variance depends on the interval length can help determine how to approximate general arrivalrate functions by piecewiselinear ones. We also develop statistical tests to determine whether the linear Poisson model is appropriate.
 Basawa, I.V., Prakasa Rao, B.L.S. (1980) Statistical Inference for Stochastic Processes. Academic Press, New York
 Brown, M. Statistical analysis of nonhomogeneous Poisson processes. In: Lewis, P.A.W. eds. (1972) Stochastic Point Processes: Statistical Analysis, Theory and Applications. Wiley, New York, pp. 6789
 Carroll, R.J. (1982) Adapting for heteroscedasticity in linear models. Ann. Statist. 4: pp. 12241233
 G.L. Choudhury, D.M. Lucantoni and W. Whitt, Numerical solution of piecewisestationary M_{t}/G_{t}/1 queues, Oper. Res., to appear.
 Cole, A.C. Progress in forecasting and traffic measurement techniques as seen at recent teletraffic congresses. In: Bonatti, M. eds. (1989) Teletraffic Science for New CostEffective Systems, Networks and Services, ITC12. Elsevier, Amsterdam, pp. 198206
 Davis, J.L., Massey, W.A., Whitt, W. (1995) Sensitivity to the servicetime distribution in the nonstationary Erlang loss model. Manag. Sci. 14: pp. 11071116
 Draper, N.R., Smith, H. (1966) Applied Regression Analysis. Wiley, New York
 Frost, V., Melamed, B. (1994) Traffic modeling for telecommunication networks. IEEE Commun. Magazine 32: pp. 7081
 Hall, R.W. (1991) Queueing Methods for Services and Manufacturing. Prentice Hall, Englewood Cliffs, NJ
 J.M. Holtzman and D.L. Jagerman, Estimating peakedness from arrival counts,Proc. 9th Int. Teletraffic Congress, Torremolinos, Spain.
 W. Leland, M. Taqqu, W. Willinger and D. Wilson, On the selfsimilar nature of ethernet traffic, IEEE/ACM Trans. Networking 2(1994).
 Massey, W.A., Whitt, W. (1993) Networks of infiniteserver queues with nonstationary Poisson input. Queueing Systems 13: pp. 183250
 W.A. Massey and W. Whitt, Stationaryprocess approximations for the nonstationary Erlang loss model, Oper. Res., to appear.
 McCullagh, P., Neider, J.A. (1983) Generalized Linear Models. Chapman and Hall, London
 MeierHellstern, K.S., Wirth, P.E., Yan, Y., Hoeflin, D.A. Traffic models for ISDN data users: office automation application. In: Jensen, A., Iversen, V.B. eds. (1991) Teletraffic and Datatraffic in a Period of Change, ITC13. Elsevier, Amsterdam, pp. 167172
 Mood, A.M., Graybill, F.A. (1963) Introduction to the Theory of Statistics. McGrawHill, New York
 Rahko, K. Measurements for control and modelling of teletraffic. In: Jensen, A., Iversen, V.B. eds. (1991) Teletraffic and Datatraffic in a Period of Change, ITC13. Elsevier, Amsterdam, pp. 609614
 Roberts, J.W. (1992) Performance Evaluation and Design of Multiservice Networks. Commission of the European Communities, Luxembourg
 Snyder, D.L., Miller, M.I. (1991) Random Point Processes in Time and Space. Springer, New York
 Taaffe, M.R., Ong, K.L. (1987) ApproximatingPh(t)/M(t)/S/C queueing systems. Ann. Oper. Res. 8: pp. 103116
 Weisberg, S. (1985) Applied Linear Regression. Wiley, New York
 Title
 Estimating the parameters of a nonhomogeneous Poisson process with linear rate
 Journal

Telecommunication Systems
Volume 5, Issue 2 , pp 361388
 Cover Date
 19960901
 DOI
 10.1007/BF02112523
 Print ISSN
 10184864
 Online ISSN
 15729451
 Publisher
 Baltzer Science Publishers, Baarn/Kluwer Academic Publishers
 Additional Links
 Topics
 Industry Sectors
 Authors

 William A. Massey ^{(1)}
 Geraldine A. Parker ^{(1)} ^{(2)}
 Ward Whitt ^{(1)} ^{(3)}
 Author Affiliations

 1. AT&T Bell Laboratories, Room 2C120, 079740636, Murray Hill, NJ, USA
 2. Educational Testing service, 08554, Princeton, NJ, USA
 3. AT&T Bell Laboratories, Room 2C178, 079740636, Murray Hill, NJ, USA