Computational Probability pp 257-323 | Cite as

# An Introduction to Numerical Transform Inversion and Its Application to Probability Models

## Abstract

Numerical transform inversion has an odd place in computational probability. Historically, transforms were exploited extensively for solving queueing and related probability models, but only rarely was numerical inversion attempted. The model descriptions were usually left in the form of transforms. Vivid examples are the queueing books by Takács [Takács, 1962] and Cohen [Cohen, 1982]. When possible, probability distributions were calculated analytically by inverting transforms, e.g., by using tables of transform pairs. Also, moments of probability distributions were computed analytically by differentiating the transforms and, occasionally, approximations were developed by applying asymptotic methods to transforms, but only rarely did anyone try to compute probability distributions by numerically inverting the available transforms. However, there were exceptions, such as the early paper by Gaver [Gaver, 1966]. (For more on the history of numerical transform inversion, see our earlier survey [Abate and Whitt, 1992a].) Hence, in the application of probability models to engineering, transforms became regarded more as mathematical toys than practical tools. Indeed, the conventional wisdom was that numerical transform inversion was very difficult. Even numerical analysts were often doubtful of the numerical stability of inversion algorithms. In queueing, both theorists and practitioners lamented about the “Laplace curtain” obscuring our understanding of system behavior.

## Keywords

Discretization Error Tail Probability Inversion Algorithm Computational Probability Numerical Inversion## Preview

Unable to display preview. Download preview PDF.

## References

- [Abate et al., 1995a]Abate, J., Choudhury, G. L., Lucantoni, D. M., and Whitt, W. (1995a). Asymptotic analysis of tail probabilities based on the computation of moments.
*Ann. Appl. Prob*., 5: 983–1007.CrossRefGoogle Scholar - [Abate et al., 1993]Abate, J., Choudhury, G. L., and Whitt, W. (1993). Calculation of the GI/G/1 waiting-time distribution and its cumulants from Pollaczek’s formulas.
*Archiv fir Elektronik and Übertragungstechnik*, 47: 31 1321.Google Scholar - [Abate et al., 1994]Abate, J., Choudhury, G. L., and Whitt, W. (1994). Waiting-time tail probabilities in queues with long-tail service-time distributions.
*Queueing Systems*, 16: 311–338.CrossRefGoogle Scholar - [Abate et al., 1995b]Abate, J., Choudhury, G. L., and Whitt, W. (1995b). Exponential approximations for tail probabilities in queues, I: waiting times.
*Oper. Res*., 43: 885–901.CrossRefGoogle Scholar - [Abate et al., 1996]Abate, J., Choudhury, G. L., and Whitt, W. (1996). On the Laguerre method for numerically inverting Laplace transforms.
*INFORMS Journal on Computing*, 8: 413–427.CrossRefGoogle Scholar - [Abate et al., 1997]Abate, J., Choudhury, G. L., and Whitt, W. (1997). Numerical inversion of multidimensional Laplace transforms by the Laguerre method.
*Performance Evaluation*, 31: 229–243.CrossRefGoogle Scholar - [Abate and Whitt, 1992a]Abate, J. and Whitt, W. (1992a). The Fourier-series method for inverting transforms of probability distributions.
*Queueing Systems*, 10: 5–88.CrossRefGoogle Scholar - [Abate and Whitt, 1992b]Abate, J. and Whitt, W. (1992b). Numerical inver- sion of probability generating functions.
*Oper. Res. Letters*, 12: 245–251.CrossRefGoogle Scholar - [Abate and Whitt, 1992c]Abate, J. and Whitt, W. (1992c). Solving probability transform functional equations for numerical inversion.
*Oper. Res. Letters*, 12: 275–281.CrossRefGoogle Scholar - [Abate and Whitt, 1995]Abate, J. and Whitt, W. (1995). Numerical inversion of Laplace transforms of probability distributions.
*ORSA J. on Computing*, 7: 36–43.CrossRefGoogle Scholar - [Abate and Whitt, 1998]Abate, J. and Whitt, W. (1998). Calculating transient characteristics of the Erlang loss model by numerical transform inversion.
*Stochastic Models*, 14.Google Scholar - [Asmussen, 1987]Asmussen, S. (1987).
*Applied Probability and Queues*. Wiley, New York.Google Scholar - [Asmussen et al., 1996]Asmussen, S., Nerman, O., and Olsson, M. (1996). Fitting phase type distributions via the EM algorithm.
*Scand. J. Statist*., 23: 419–441.Google Scholar - [Bene, 1961]Bene, V. E. (1961). The covariance function of a simple trunk group with applications to traffic measurements.
*Bell System Tech. J*., 40: 117–148.Google Scholar - [Benes, 1965]Benes, V. E. (1965).
*Mathematical Theory of Connecting Networks and Telephone Traffic*. Academic Press, New York.Google Scholar - [Bertozzi and McKenna, 1993]Bertozzi, A. and McKenna, J. (1993). Multidimensional residues, generating functions, and their application to queueing networks.
*SIAM Review*, 35: 239–268.CrossRefGoogle Scholar - [Borovkov, 1984]Borovkov, A. A. (1984).
*Asymptotics Methods in Queueing Theory*. Wiley, New York.Google Scholar - [Chaudhry et al., 1992]Chaudhry, M. L., Agarwal, M., and Templeton, J. G. C. (1992). Exact and approximate numerical solutions of steady-state distributions arising in the GI/G/1 queue.
*Queueing Systems*, 10: 105–152.CrossRefGoogle Scholar - [Choudhury et al., 1995a]Choudhury, G. L., Leung, K. K., and Whitt, W. (1995a). An algorithm to compute blocking probabilities in multi-rate multi-class multi-resource loss models. Adv.
*Appl. Prob*., 27: 1104–1143.CrossRefGoogle Scholar - [Choudhury et al., 1995b]Choudhury, G. L., Leung, K. K., and Whitt, W. (1995b). Calculating normalization constants of closed queueing networks by numerically inverting their generating functions.
*J. ACM*, 42: 935–970.CrossRefGoogle Scholar - [Choudhury et al., 1995c]Choudhury, G. L., Leung, K. K., and Whitt, W. (1995c). Efficiently providing multiple grades of service with protection against overloads in shared resources.
*ATandT Tech. J*., 74: 50–63.Google Scholar - [Choudhury et al., 1995d]Choudhury, G. L., Leung, K. K., and Whitt, W. (1995d). An inversion algorithm to compute blocking probabilities in loss networks with state-dependent rates.
*IEEE/ACM Trans. Networking*, 3: 585601Google Scholar - [Choudhury and Lucantoni, 1996]Choudhury, G. L. and Lucantoni, D. M. (1996). Numerical computation of the moments of a probability distribution from its transform.
*Oper. Res*., 44: 368–381.CrossRefGoogle Scholar - [Choudhury et al., 1994]Choudhury, G. L., Lucantoni, D. M., and Whitt, W. (1994). Multidimensional transform inversion with applications to the transient M/G/1 queue. Ann.
*Appl. Prob*., 4: 719–740.CrossRefGoogle Scholar - [Choudhury et al., 1996]Choudhury, G. L., Lucantoni, D. M., and Whitt, W. (1996). Squeezing the most out of ATM.
*IEEE Trans. Commun*., 44: 203–217.CrossRefGoogle Scholar - [Choudhury et al., 1997]Choudhury, G. L., Lucantoni, D. M., and Whitt, W. (1997). Numerical solution of Mt/Gt/1 queues.
*Oper. Res*., 45: 451–463.CrossRefGoogle Scholar - Choudhury and Whitt, 1995] Choudhury, G. L. and Whitt, W. (1995). Q
^{2}: A new performance analysis tool exploiting numerical transform inversion. In*Proc. Third Int. Workshop on Modeling, Analysis and Simul. of Computer and Telecomm. Systems*,pages 411–415, Durham, NC. (MASCOTS ’85).Google Scholar - [Choudhury and Whitt, 1996]Choudhury, G. L. and Whitt, W. (1996). Computing distributions and moments in polling models by numerical transform inversion.
*Performance Evaluation*, 25: 267–292.CrossRefGoogle Scholar - [Choudhury and Whitt, 1997]Choudhury, G. L. and Whitt, W. (1997). Probabilistic scaling for the numerical inversion of non-probability transforms.
*INFORMS J. Computing*, 9: 175–184.CrossRefGoogle Scholar - [Cohen, 1982]Cohen, J. W. (1982).
*The Single Server Queue*. North-Holland, Amsterdam, second edition.Google Scholar - [Conway and Georganas, 1989]Conway, A. E. and Georganas, N. D. (1989).
*Queueing Networks — Exact Computational Algorithms: A Unified Theory Based on Decomposition and Aggregation*. MIT Press, Cambridge, MA.Google Scholar - [Darling and Siegert, 1953]Darling, D. A. and Siegert, J. F. (1953). The first passage problem for a continuous Markov process.
*Ann. Math. Statist*., 24: 624–639.CrossRefGoogle Scholar - [Davies and Martin, 1979]Davies, B. and Martin, B. L. (1979). Numerical inversion of the Laplace transform: A survey and comparison of methods.
*J. Comp. Phys*., 33: 1–32.CrossRefGoogle Scholar - [Davis and Rabinowitz, 1984]Davis, P. J. and Rabinowitz, P. (1984).
*Methods of Numerical Integration*. Academic Press, New York, second edition.Google Scholar - [Doetsch, 1961]Doetsch, G. (1961).
*Guide to Applications of Laplace Transforms*. Van Nostrand, London.Google Scholar - [Doetsch, 1974]Doetsch, G. (1974).
*Introduction to the Theory and Application of the Laplace Transformation*. Springer-Verlag, New York.CrossRefGoogle Scholar - [Dubner and Abate, 1968]Dubner, H. and Abate, J. (1968). Numerical inversion of Laplace transforms by relating them to the finite Fourier cosine transform.
*J. ACM*, 15: 115–123.CrossRefGoogle Scholar - [Duffield and Whitt, 1998]Duffield, N. G. and Whitt, W. (1998). A source traffic model and its transient analysis for network control.
*Stochastic Models*, 14.Google Scholar - [Durbin, 1974]Durbin, F. (1974). Numerical inversion of Laplace transforms: an efficient improvement to Dubner and Abate’s method.
*Comput. J*., 17: 371–376.CrossRefGoogle Scholar - [Feldmann and Whitt, 1997]Feldmann, A. and Whitt, W. (1997). Fitting mixtures of exponentials to long-tail distributions to analyze network performance models.
*Performance Evaluation*, 31: 245–279.CrossRefGoogle Scholar - [Feller, 1971]Feller, W. (1971).
*An Introduction to Probability Theory and its Applications*, volume II. Wiley, New York, second edition.Google Scholar - [Gaver, 1966]Gaver, D. P. (1966). Observing stochastic processes and approximate transform inversion.
*Operations Research*, 14: 444–459.CrossRefGoogle Scholar - [Gaver and Jacobs, 1998]Gaver, D. P. and Jacobs, P. A. (1998). Waiting times when service times are stable laws: tamed and wild. In Shanthikumar, J. G. and (eds.), U. S., editors,
*Recent Contributions in Applied Probability and Stochastic Processes, Festschrift for Julian Keilson*. Kluwer, Boston.Google Scholar - [Giffin, 1975]Giffin, W. C. (1975).
*Transform Techniques for Probability Modeling*. Academic Press, New York.Google Scholar - [Hosono, 1979]Hosono, T. (1979). Numerical inversion of Laplace transform.
*J. Inst. Elec. Eng. Jpn*., pages A54 - A64. 494 (In Japanese).Google Scholar - [Hosono, 1981]Hosono, T. (1981). Numerical inversion of Laplace transform and some applications to wave optics.
*Radio Sci*., 16: 1015–1019.CrossRefGoogle Scholar - Hosono, 1984] Hosono, T. (1984).
*Fast Inversion of Laplace Transform by BASIC*. Kyoritsu Publishers, Japan. (In Japanese).Google Scholar - [Jagerman, 1974]Jagerman, D. L. (1974). Some properties of the Erlang loss function.
*Bell System Tech. J*., 53: 525–551.Google Scholar - [Jagerman, 1978]Jagerman, D. L. (1978). An inversion technique for the Laplace transform with applications.
*Bell System Tech. J*., 57: 669–710.Google Scholar - [Jagerman, 1982]Jagerman, D. L. (1982). An inversion technique for the Laplace transform.
*Bell Sys. Tech. J*., 61: 1995–2002.Google Scholar - [Johnsonbaugh, 1979]Johnsonbaugh, R. (1979). Summing an alternating series.
*Amer. Math. Monthly*, 86: 637–648.CrossRefGoogle Scholar - [Kao, 1997]Kao, E. P. C. (1997).
*An Introduction to Stochastic Processes*. Duxbury Press, New York.Google Scholar - [Keilson, 1979]Keilson, J. (1979).
*Markov Chain Models — Rarity and Exponentiality*. Springer-Verlag, New York.CrossRefGoogle Scholar - [Keilson and Ross, 1975]Keilson, J. and Ross, H. F. (1975). Passage time distributions for Gaussian Markov (Ornstein-Uhlenbech) statistical processes.
*Selected Tables in Mathematical Statistics*, 3: 233–327.Google Scholar - [Kingman, 1961]Kingman, J. F. C. (1961). The single server queue in heavy traffic.
*Proc. Camb. Phil. Soc*., 57: 902–904.CrossRefGoogle Scholar - [Kleinrock, 1975]Kleinrock, L. (1975).
*Queueing Systems, Volume I: Theory*. Wiley, New York.Google Scholar - [Knessl, 1990]Knessl, C. (1990). On the transient behavior of the M/M/m/m loss model.
*Stochastic Models*, 6: 749–776.CrossRefGoogle Scholar - [Kobayashi, 1978]Kobayashi, H. (1978).
*Modeling and Analysis: An Introduction to System Performance Evaluation Methodology*. Addison-Wesley, Reading, MA.Google Scholar - [Kwok and Barthez, 1989]Kwok, Y. K. and Barthez, D. (1989). An algorithm for the numerical inversion of the Laplace transform.
*Inverse Problems*, 5: 1089–1095.CrossRefGoogle Scholar - [Lam and Lien, 1983]Lam, S. S. and Lien, Y. L. (1983). A tree convolution algorithm for the solution of queueing networks.
*Commun. ACM*, 26: 203215.Google Scholar - [Lavenberg, 1983]Lavenberg, S. S., editor (1983).
*Computer Performance Modeling Handbook*. Academic Press, Orlando, FL.Google Scholar - [Lucantoni et al., 1994]Lucantoni, D. M., Choudhury, G. L., and Whitt, W. (1994). The transient BMAP/G/1 queue.
*Stochastic Models*, 10: 145–182.CrossRefGoogle Scholar - [Mitra and Weiss, 1989]Mitra, D. and Weiss, A. (1989). The transient behavior in Erlang’s model for large trunk groups and various traffic conditions. In
*Teletraffic Science for new cost-Effective Systems, Networks and Services*, pages 1367–1374, Amsterdam. Elsevier-Science.Google Scholar - [Neuts, 1981]Neuts, M. F. (1981).
*Matrix-Geometric Solutions in Stochastic Models*. The Johns Hopkins University Press, Baltimore.Google Scholar - [Neuts, 1989]Neuts, M. F. (1989).
*Structured Stochastic Matrices of M/G/1 Type and Their Applications*. Marcel Dekker, New York.Google Scholar - [O’Cinneide, 1997]O’Cinneide, C. A. (1997). Euler summation for Fourier series and Laplace transform inversion.
*Stochastic Models*, to appear.Google Scholar - [Pol and Bremmer, 1987]Pol, B. V. D. and Bremmer, H. (1987).
*Operational Calculus*. Cambridge Press, reprinted Chelsea Press, New York.Google Scholar - [Pollaczek, 1952]Pollaczek, F. (1952). Fonctions caractéristiques de certaines répartitions définies au money de la notion d’ordre. Application à la théorie de attentes.
*C. R. Acad. Sci. Paris*, 234: 2334–2336.Google Scholar - [Pollaczek, 1965]Pollaczek, F. (1965). Concerning an analytic method for the treatment of queueing problems. In Smith, W. L. and Wilkinson, W. E., editors,
*Proceedings of the Symposium on Congestion Theory*, pages 1–25 and 34–42, Chapel Hill. The University of North Carolina Press.Google Scholar - [Poularikas, 1996]Poularikas, A. D. (1996).
*The Transforms and Applications Handbook*. CRC Press, Boca Raton, FL.Google Scholar - [Press et al., 1988]Press, W. H., Flannery, B. P., Teukolsky, S. A., and Vetterling, W. T. (1988).
*Numerical Recipes, FORTRAN Version*. Cambridge University Press, Cambridge, England.Google Scholar - [Reif, 1965]Reif, F. (1965).
*Fundamentals of Statistical and Thermal Physics*. McGraw-Hill, New York.Google Scholar - [Reiser and Kobayashi, 1975]Reiser, M. and Kobayashi, H. (1975). Queueing networks with multiple closed chains: theory and computational algorithms.
*IBM J. Res. Dev*., 19: 283–294.CrossRefGoogle Scholar - [Riordan, 1962]Riordan, J. (1962).
*Stochastic Service Systems*. Wiley, New York.Google Scholar - [Simon et al., 1972]Simon, R. M., Stroot, M. T., and Weiss, G. H. (1972). Numerical inversion of Laplace transforms with applications to percentage labeled experiments.
*Comput. Biomed. Res*., 6: 596–607.CrossRefGoogle Scholar - [Smith, 1953]Smith, W. L. (1953). On the distribution of queueing times.
*Proc. Camb. Phil. Soc*., 49: 449–461.CrossRefGoogle Scholar - [Srikant and Whitt, 1996]Srikant, R. and Whitt, W. (1996). Simulation run lengths to estimate blocking probabilities.
*ACM J. TOMACS*, 6: 7–52.CrossRefGoogle Scholar - [Takâcs, 1962]Takâcs, L. (1962).
*Introduction to the Theory of Queues*. Oxford University Press, New York.Google Scholar - [Tolstov, 1976]Tolstov, G. P. (1976).
*Fourier Series*. Dover, New York.Google Scholar - [Weeks, 1966]Weeks, W. T. (1966). Numerical inversion of Laplace transforms using Laguerre functions.
*J. ACM*, 13: 419–426.CrossRefGoogle Scholar - [Whitt, 1984]Whitt, W. (1984). Heavy-traffic approximations for service systems with blocking.
*ATandT Bell Lab. Tech. J*., 63: 689–708.Google Scholar - [Wimp, 1981]Wimp, J. (1981).
*Sequence Transformations and Their Applications*. Academic Press, New York.Google Scholar