Abstract
Simulation is commonly used to study the random behaviors of large-scale stochastic systems with correlated inputs. Since the input correlation is often induced by latent common factors in many situations, to facilitate system diagnostics and risk management, we introduce a factor-based Bayesian framework that can improve both computational and statistical efficiency and provide insights for system risk analysis. Specifically, we develop a flexible Gaussian copula-based multivariate input model that can capture important properties in the real-world data. A nonparametric Bayesian approach is used to model marginal distributions, and it can capture the properties, including multi-modality and skewness. We explore the factor structure of the underlying generative processes for the dependence. Both input and simulation estimation uncertainty are characterized by the posterior distributions. In addition, we interpret the latent factors and estimate their effects on the system performance, which could be used to support diagnostics and decision making for large-scale stochastic systems. Our approach is supported by both asymptotic theory and empirical study.
- O. Aguilar and M. West. 2000. Bayesian dynamic factor models and portfolio allocation. Journal of Business 8 Economic Statistics 18, 3 (2000), 338--357. http://www.jstor.org/stable/1392266.Google Scholar
- A. Akcay, B. Biller, and S. Tayur. 2011. Improved inventory targets in the presence of limited historical demand data. Manufact. Serv. Operat. Manage. 13, 3 (2011), 297--309. Google ScholarDigital Library
- Adelchi Azzalini and Antonella Capitanio. 2003. Distributions generated by perturbation of symmetry with emphasis on a multivariate skew t-distribution. J. Roy. Stat. Soc. B 65, 2 (2003), 367--389. http://www.jstor.org/stable/3647510Google ScholarCross Ref
- A. Azzalini, A. with the collaboration of Capitanio. 2014. The Skew-normal and Related Families. Cambridge University Press, Cambridge, UK.Google Scholar
- R. R. Barton. 2007. Presenting a more complete characterization of uncertainty: Can it be done? In Proceedings of the 2007 INFORMS Simulation Society Research Workshop.Google Scholar
- R. R. Barton. 2012. Tutorial: Input uncertainty in output analysis. In Proceedings of the 2012 Winter Simulation Conference, C. Laroque, J. Himmelspach, R. Pasupathy, O. Rose, and A. M. Uhrmacher (Ed.). IEEE, 67--78. Google ScholarDigital Library
- R. R. Barton, B. L. Nelson, and W. Xie. 2014. Quantifying input uncertainty via simulation confidence intervals. Informs J. Comput. 26 (2014), 74--87. Google ScholarDigital Library
- R. R. Barton and L. W. Schruben. 1993. Uniform and bootstrap resampling of input distributions. In Proceedings of the 1993 Winter Simulation Conference. Institute of Electrical and Electronics Engineers, Inc., Piscataway, NJ, 503--508. Google ScholarDigital Library
- B. Biller, A. Akcay, C. Corlu, and S. Tayur. 2014. A simulation-based support tool for data-driven decision making: Operational testing for dependence modeling. In Proceedings of the 2014 Winter Simulation Conference (WSC’14). 899--909. Google ScholarDigital Library
- B. Biller and C. G. Corlu. 2011. Accounting for parameter uncertainty in large-scale stochastic simulations with correlated inputs. Operat. Res. 59 (2011), 661--673. Google ScholarDigital Library
- B. Biller and C. G. Corlu. 2012. Copula-based multivariate input modeling. Surv. Operat. Res. Manage. Sci. 17 (2012), 69--84.Google ScholarCross Ref
- B. Biller and S. Ghosh. 2006. Multivariate input processes. In Handbooks in Operations Research and Management Science: Simulation, S. Henderson and B. L. Nelson (Eds.). Elsevier, Chapter 5.Google Scholar
- B. Biller and C. Gunes. 2010. Introduction to simulation input modeling. In Proceedings of the 2010 Winter Simulation Conference, B. Johansson, S. Jain, J. Montoya-Torres, J. Hugan, and E. Yucesan (Eds.). Institute of Electrical and Electronics Engineers, Inc., Piscataway, NJ. Google ScholarDigital Library
- B. Biller and B. L. Nelson. 2009. Modeling and generating multivariate time-series input processes using a vector autoregressive technique. ACM Trans. Model. Comput. Simul. 13, 3 (2009), 211--237. Google ScholarDigital Library
- C. M. Bishop. 2006. Pattern Recognition and Machine Learning. Springer, New York. Google ScholarDigital Library
- M. C. Cario and B. L. Nelson. 1997. Modeling and Generating Random Vectors with Arbitrary Marginal Distributions and Correlation Matrix. Technical Report. Department of Industrial Engineering and Management Sciences, Northwestern University.Google Scholar
- R. C. H. Cheng and W. Holland. 1997. Sensitivity of computer simulation experiments to errors in input data. J. Stat. Comput. Simul. 57 (1997), 219--241.Google ScholarCross Ref
- R. C. H. Cheng and W. Holland. 2004. Calculation of confidence intervals for simulation output. ACM Trans. Model. Comput. Simul. 14 (2004), 344--362. Google ScholarDigital Library
- S. E. Chick. 2001. Input distribution selection for simulation experiments: Accounting for input uncertainty. Operat. Res. 49 (2001), 744--758. Google ScholarDigital Library
- W. J. Conover. 1980. Practical Nonparametric Statistics. John Wiley 8 Sons, New York, NY.Google Scholar
- M. Drton. 2009. Likelihood ratio tests and singularities. The Annals of Statistics 37, 2 (2009), 979--1012.Google ScholarCross Ref
- M. Drton and M. Plummer. 2013. A bayesian information criterion for singular models. Journal of the Royal Statistical Society: Series B (Statistical Methodology) 79, 2 (2017), 323--380.Google ScholarCross Ref
- M. Drton, B. Sturmfels, and S. Sullivant. 2007. Algebraic factor analysis: Tetrads, pentads and beyond. Probability Theory and Related Fields 138, 3 (2007), 463--493.Google ScholarCross Ref
- D. B. Dunson and J. A. Taylor. 2005. Approximate bayesian inference for quantiles. Journal of Nonparametric Statistics 17, 3 (2005), 385--400.Google ScholarCross Ref
- M. D. Escobar and M. West. 1995. Bayesian density estimation and inference using mixtures. J. Am. Stat. Assoc. 90, 430 (1995), 577--588.Google ScholarCross Ref
- A. Gelman, J. B. Carlin, H. S. Stern, and D. B. Rubin. 2004. Bayesian Data Analysis (2nd ed.). Taylor 8 Francis Group, LLC, New York, NY.Google Scholar
- J. F. Geweke and K. J. Singleton. 1980. Interpreting the likelihood ratio statistic in factor models when sample size is small. J. Am. Stat. Assoc. 75, 369 (1980), 133--137.Google ScholarCross Ref
- S. Ghosal, J. K. Ghosh, and R. V. Ramamoorthi. 1999. Posterior consistency of dirichlet mixtures in density estimation. Ann. Stat. 27, 1 (1999), 143--158.Google ScholarCross Ref
- S. Ghosal and A. W. van der Vaart. 2007. Posterior convergence rates of dirichlet mixtures at smooth densities. Ann. Stat. 35, 2 (2007), 697--723.Google ScholarCross Ref
- J. K. Ghosh and R. V. Ramamoorthi. 2003. Bayesian Nonparametrics. Springer-Verlag, New York, NY.Google Scholar
- S. Ghosh and S. G. Henderson. 2002. Properties of the NORTA method in higher dimensions. In Proceedings of the 2002 Winter Simulation Conference, E. Yűcesan, C. H. Chen, J. L. Snowdon, and J. M. Charnes (Eds.). Institute of Electrical and Electronics Engineers, Inc., Piscataway, NJ, 263--269. Google ScholarDigital Library
- R. J. Hanson. 1975. Stably updating mean and standard deviation of data. Commun. ACM 18, 1 (Jan. 1975), 57--58. Google ScholarDigital Library
- P. D. Hoff. 2007. Extending the rank likelihood for semiparametric copula estimation. Ann. Appl. Stat. 1 (2007), 265--283.Google ScholarCross Ref
- L. J. Hong, Z. Hu, and G. Liu. 2014. Monte carlo methods for value-at-risk and conditional value-at-risk: A review. ACM Trans. Model. Comput. Simul. 24, 4, Article 22 (2014), 37 pages. Google ScholarDigital Library
- K. Horiguchi, N. Raghavan, R. Uzsoy, and S. Venkateswaran. 2001. Finite-capacity production planning algorithms for a semiconductor wafer fabrication facility. Int. J. Prod. Res. 39 (2001), 825--842.Google ScholarCross Ref
- J. C. Hsu and B. L. Nelson. 1990. Control variate for quantile estimation. Manage. Sci. 36, 7 (1990), 835--851. Google ScholarDigital Library
- H. Jeffreys. 1961. Theory of Probability (3 ed.). Claredon, Oxford.Google Scholar
- H. Joe. 2005. Asymptotic efficiency of the two-stage estimation method for copula-based models. J. Multivar. Anal. 94 (2005), 401419. Google ScholarDigital Library
- H. Lam. 2016. Advanced tutorial: Input uncertainty and robust analysis in stochastic simulation. In Proceedings of the 2016 Winter Simulation Conference, T. M. K. Roeder, P. I. Frazier, R. Szechtman, E. Zhou, T. Huschka, and S. E. Chick (Eds.). Institute of Electrical and Electronics Engineers, Inc., Piscataway, NJ. Google ScholarDigital Library
- T. Lancaster and J. J. Sung. 2010. Bayesian quantile regression methods. J. Appl. Econometr. 25, 2 (2010), 287--307.Google ScholarCross Ref
- M. Lavine. 1995. On an approximate likelihood for quantiles. Biometrika 82, 1 (1995), 220--222.Google ScholarCross Ref
- H. F. Lopes and M. West. 2004. Bayesian model assessment in factor analysis. Stat. Sin. 14 (2004), 41--67.Google Scholar
- Y. Ma. 2011. Risk Management in Biopharmaceutical Supply Chains. Ph.D. Dissertation. University of California, Berkeley.Google Scholar
- B. Masih-Tehrani, S. H. Xu, S. Kumara, and H. Li. 2011. A single-period analysis of a two-echelon inventory system with dependent supply uncertainty. Transport. Res. 45, 1128--1151 (2011).Google ScholarCross Ref
- J. S. Murray, D. B. Dunson, L. Carin, and J. E. Lucas. 2013. Bayesian gaussian copula factor models for mixed data. Journal of the American Statistical Association 108, 502 (2013), 656--665.Google ScholarCross Ref
- B. L. Nelson. 2016. ‘Some tactical problems in digital simulation’ for the next 10 years. Journal of Simulation 10, 1 (2016), 2--11.Google ScholarCross Ref
- S. H. Ng and S. E. Chick. 2006. Reducing parameter uncertainty for stochastic systems. ACM Transactions on Modeling and Computer Simulation 16, 1 (2006), 26--51. Google ScholarDigital Library
- T. A. Severini. 2000. Likelihood Methods in Statistic. Oxford Statistical Science Series. Oxford University Press, Oxford.Google Scholar
- M. S. Smith. 2011. Bayesian Approaches to Copula Modeling. Technical Report. University of Melbourne.Google Scholar
- E. Song, B. L. Nelson, and C. D. Pegden. 2014. Advanced tutorial: Input uncertainty quantification. In Proceedings of the 2014 Winter Simulation Conference, A. Tolk, S. Y. Diallo, I. O. Ryzhov, L. Yilmaz, S. Buckley, and J. A. Miller (Eds.). Institute of Electrical and Electronics Engineers, Inc., Piscataway, NJ. Google ScholarDigital Library
- S. M. Wagner, C. Bode, and P. Koziol. 2009. Supplier default dependencies: Empirical evidence from the automotive industry. European Journal of Operational Research 199, 1 (2009), 150--161.Google ScholarCross Ref
- M. West. 1990. Bayesian Kernel Density Estimation. Technical Report. Duke University.Google Scholar
- W. Xie, C. Li, and P. Zhang. 2017. A Bayesian nonparametric hierarchical framework for uncertainty quantification in simulation. (2017). under review.Google Scholar
- W. Xie, B. L. Nelson, and R. R. Barton. 2014. A bayesian framework for quantifying uncertainty in stochastic simulation. Operat. Res. 62, 6 (2014), 1439--1452.Google ScholarDigital Library
- W. Xie, B. L. Nelson, and R. R. Barton. 2016. Multivariate input uncertainty in output analysis for stochastic simulation. ACM Trans. Model. Comput. Simul. 27, 5 (2016). Google ScholarDigital Library
- F. Zouaoui and J. R. Wilson. 2003. Accounting for parameter uncertainty in simulation input modeling. IIE Trans. 35 (2003), 781--792.Google ScholarCross Ref
- F. Zouaoui and J. R. Wilson. 2004. Accounting for input-model and input-parameter uncertainties in simulation. IIE Trans. 36 (2004), 1135--1151.Google ScholarCross Ref
Index Terms
- A Factor-Based Bayesian Framework for Risk Analysis in Stochastic Simulations
Recommendations
A generic framework for stochastic Loss-Given-Default
In this document a method is discussed to incorporate stochastic Loss-Given-Default (LGD) in factor models, i.e. structural models for credit risk. The general idea exhibited in this text is to introduce a common dependence of the LGD and the ...
Robust sparse Bayesian infinite factor models
AbstractMost of previous works and applications of Bayesian factor model have assumed the normal likelihood regardless of its validity. We propose a Bayesian factor model for heavy-tailed high-dimensional data based on multivariate Student-t likelihood to ...
Nonparametric methods for unbalanced multivariate data and many factor levels
We propose different nonparametric tests for multivariate data and derive their asymptotic distribution for unbalanced designs in which the number of factor levels tends to infinity (large a, small n"i case). Quasi gratis, some new parametric ...
Comments