Sage Journals: Discover world-class research

Abstract

The generation of correlated random variates requires a positive definite correlation matrix. This paper presents an interactive approach to generating a positive definite correlation matrix from subjective information. The procedure, implemented in a FORTRAN routine, facilitates assessment of correlation structures determined from subjective considerations by providing feedback on the feasibility of assumed correlation structures. The routine was verified by applying it to a positive definite correlation matrix reported in the literature. An application of the routine to a nonpositive definite correlation matrix used in a capital investiment simulation for a new papermaking technology illustrates how the procedure may be used to assess possible correlation structures with subjective information.

Keywords

Monte Carlo simulation modeling correlation matrix subjective information

Get full access to this article

View all access options for this article.

References

Brown, B.W. , and J. Lovato. 1991. RANLIB. Library of FORTRAN routines for random number generation. Dept. of Biomathematics. Univ. of Texas, M. D. Anderson Cancer Center, Houston, Texas.

Carsel, R.F. , and R.S. Parrish. 1988. Developing joint probability distributions of soil water retention properties. Water Resources Research 24(5):755-769.

Han, M.B. , D.A. Bender , and S.E. Taylor. 1991. Computer generation of highly skewed correlated random variables. Trans. of the ASAE 34(5):2279-2281.

Imsl. 1987. Stat/math/sfun Libraries - FORTRAN subroutines for statistical and mathematical analysis. International Mathematical and Statistical Library, Houston, Texas.

Hu, H. , and I. Olkin. 1991. A numerical procedure for finding the positive definite matrix closest to a patterned matrix. Statistics and Probability Letters 12:511-515.

Ince, P.J. , and J. Buongiorno. 1991. Multivariate stochastic simulation with subjective multivariate normal distributions. In: Proceedings of the 1991 symposium on systems analysis in forest resources; March 3-6, Charleston, South Carolina: U. S. Department of Agriculture, Forest Service, Southeastern Forest Experiment Station, 1991:143-150.

Law, A.M. , and W.D. Kelton. 1991. Simulation modeling and analysis. McGraw-Hill Inc., New York.

Microsoft. 1991. Microsoft FORTRAN Version 5.1. Users reference. Redmond, WA: Microsoft Corporation.

Oren, S.S. 1981. A conjugate directions method for subjective assessment of normal distributions. Journal of Optimization Theory and Applications 33(1):25-3H.

10.

Scheuer, E.M. , and D.S. Stoller. 1962. On the generation of normal random vectors . Technometrics 4:278-281.

11.

Song, Q. , and L.C. Brown. 1990. DO model uncertainty with correlated inputs. Journal of EnvironmentalEngineering 116(6):1164-1180.

12.

Taylor, S.E. and D.A. Bender. 1989. A method for simulating multiple correlated lumber properties. Forest Products journal 39(7/8):71-74.

Assessing Correlation Structures with Subjective Information-An Interactive Approach

Abstract

Keywords

Get full access to this article

References