Sensitivity and scenario analysis for simulation metamodels

ABSTRACT

We use simple orthogonal and non-orthogonal designs to analyze a multi-tiered model for forecasting performance of a large-scale home mortgage portfolio. The experiments are used to assess the sensitivity of performance to projected changes in economic conditions as well as the sensitivity of the model to coefficients estimated from historical data. Our results attribute the variation in loan performance to variation in individual factors or factor combinations, indicating which are crucial to monitor or forecast accurately. The results are at times counter-intuitive, indicating the benefits of a systematic approach to sensitivity assessment and scenario generation.

REFERENCES

• 1: Myers, R. H., A. I. Khuri and G. Vining. 1992. Response surface alternatives to the Taguchi robust parameter design approach. The American Statistician 46(2): 131-139.
• 2: Susan M. Sanchez, Experimental designs for system assessment and improvement when noise factors are correlated, Proceedings of the 26th conference on Winter simulation, p.290-296, December 11-14, 1994, Orlando, Florida, United States
• 3: Susan M. Sanchez, A robust design tutorial, Proceedings of the 26th conference on Winter simulation, p.106-113, December 11-14, 1994, Orlando, Florida, United States
• 4: Sanchez, S. M., P. J. Sanchez and J. S. Ramberg. 1996. A simulation framework for robust system design. In Concurrent design of products, manufacturing processes and systems, ed. B. Wang. New York: Gordon and Breach, forthcoming.
• 5: Smith, L. D., S. M. Sanchez and E. C. Lawrence. 1996. A comprehensive model for managing credit risk and forecasting losses on home mortgage portfolios. Decision Sciences, forthcoming.
• 6: Taguchi, G. 1986. Introduction to quality engineering, White Plains, New York: UNIPUB/Krauss International.
• 7: Taguchi, G. 1987. System of Experimental Design, Vols. 1 and 2. White Plains, New York: UNIPUB/Krauss International.
• 8: Paul Zipkin, Mortgages and Markov chains: a simplified evaluation model, Management Science, v.39 n.6, p.683-691, June 1993 [doi>10.1287/mnsc.39.6.683]