Kybernetika 50 no. 6, 1049-1064, 2014

Scenario generation with distribution functions and correlations

Michal Kaut and Arnt-Gunnar LiumDOI: 10.14736/kyb-2014-6-1049

Abstract:

In this paper, we present a method for generating scenarios for two-stage stochastic programs, using multivariate distributions specified by their marginal distributions and the correlation matrix. The margins are described by their cumulative distribution functions and we allow each margin to be of different type. We demonstrate the method on a model from stochastic service network design and show that it improves the stability of the scenario-generation process, compared to both sampling and a method that matches moments and correlations.

Keywords:

stochastic programming, scenario generation, moment matching, distribution functions, service network design

Classification:

90C15, 62G30, 62H20

paper.pdf

References:

  1. J. R. Birge and F. Louveaux: Introduction to stochastic programming. Springer-Verlag, New York 1997.   CrossRef
  2. J. Dupačov{á}, G. Consigli and S. W. Wallace: Scenarios for multistage stochastic programs. Ann. Oper. Res. 100 (2000), 1 - 4, 25-53. DOI: 10.1023/A:1019206915174.   CrossRef
  3. J. Dupačová, N. Gr{ö}we-Kuska and W. R{ö}misch: Scenario reduction in stochastic programming: An approach using probability metrics. Math. Programming 95 (2003), 3, 493-511. DOI: 10.1007/s10107-002-0331-0.   CrossRef
  4. S.-E. Fleten and E. Pettersen: Constructing bidding curves for a price-taking retailer in the norwegian electricity market. IEEE Trans. Power Systems 20 (2005), 2, 701-708. DOI: 10.1109/TPWRS.2005.846082.   CrossRef
  5. A. Geyer, M. Hanke and A. Weissensteiner: No-arbitrage conditions, scenario trees, and multi-asset financial optimization. European J. Oper. Res. 206 (2010), 3, 609-613. DOI: 10.1016/j.ejor.2010.03.022.   CrossRef
  6. H. Heitsch and W. R{ö}misch: Scenario reduction algorithms in stochastic programming. Comput. Optim. Appl. 24 (2003), 2 - 3, 187-206. DOI: 10.1023/A:1021805924152.   CrossRef
  7. H. Heitsch and W. R{ö}misch: Scenario tree modelling for multistage stochastic programs. Math. Programming 118 (2009), 2, 371-406. DOI: 10.1007/s10107-007-0197-2.   CrossRef
  8. H. Heitsch, W. R{ö}misch and C. Strugarek: Stability of multistage stochastic programs. SIAM J. Optim. 17 (2006), 2, 511-525. DOI: 10.1137/050632865.   CrossRef
  9. R. Henrion, C. K{ü}chler and W. R{ö}misch: Scenario reduction in stochastic programming with respect to discrepancy distances. Comput. Optim. Appl. 43 (2009), 1, 67-93. DOI: 10.1007/s10589-007-9123-z.   CrossRef
  10. K. Høyland and S. W. Wallace: Generating scenario trees for multistage decision problems. Management Sci. 47 (2001), 2, 295-307. DOI: 10.1023/A:1021853807313.   CrossRef
  11. K. Høyland, M. Kaut and S. W. Wallace: A heuristic for moment-matching scenario generation. Comput. Optim. Appl. 24 (2003), 2 - 3, 169-185.   CrossRef
  12. P. Kall and S. W. Wallace: Stochastic Programming. John Wiley and Sons, Chichester 1994.   CrossRef
  13. M. Kaut and S. W. Wallace: Evaluation of scenario-generation methods for stochastic programming. Pacific J. Optim. 3 (2007), 2, 257-271.   CrossRef
  14. M. Kaut and S. W. Wallace: Shape-based scenario generation using copulas. Comput. Management Sci. 8 (2011), 1 - 2, 181-199. DOI: 10.1007/s10287-009-0110-y.   CrossRef
  15. A.-G. Lium and M. Kaut: Scenario generation for obtaining sound solutions. In: Stochastic Service Network Design 2 (2006), Chapter 4.   CrossRef
  16. W. Mak, D. Morton and R. Wood: Monte Carlo bounding techniques for determining solution quality in stochastic programs. Oper. Res. Letters 24 (1999), 47-56.   CrossRef
  17. J. Okunev and D. R. White: Moment matching for the masses. Available at SSRN: \url{http://ssrn.com/abstract=921451}, 2006.   CrossRef
  18. G. C. Pflug: Scenario tree generation for multiperiod financial optimization by optimal discretization. Math. Programming 89 (2001), 2, 251-271. DOI: 10.1007/PL00011398.   CrossRef
  19. P. Sch{ü}tz and A. Tomasgard: The impact of flexibility on operational supply chain planning. Int. J. Production Economics 134 (2011), 2, 300-311. DOI: 10.1016/j.ijpe.2009.11.004.   CrossRef
  20. P. Sch{ü}tz, A. Tomasgard and S. Ahmed: Supply chain design under uncertainty using sample average approximation and dual decomposition. European J. Oper. Res. 199 (2009), 2 409-419. DOI: 10.1016/j.ejor.2008.11.040.   CrossRef
  21. B. K. Thapalia, T. G. Crainic, M. Kaut and S. W. Wallace: Single-commodity stochastic network design with multiple sources and sinks. Inform. Systems Oper. Res. 49 (2011), 3, 193-211. DOI: 10.3138/infor.49.3.003.   CrossRef
  22. N. Topaloglou, H. Vladimirou and S. A. Zenios: A dynamic stochastic programming model for international portfolio management. European J. Oper. Res. 185 (2008), 3, 1501-1524. DOI: 10.1016/j.ejor.2005.07.035.   CrossRef
  23. A. Werner, K. T. Uggen, M. Fodstad, A.-G. Lium and R. Egging: Stochastic mixed-integer programming for integrated portfolio planning in the {LNG} supply chain. The Energy J. 35 (2014), 1. DOI:10.5547/01956574.35.1.5.   CrossRef