Properties of the NORTA method in higher dimensions

The NORTA method for multivariate generation is a fast general purpose method for generating samples of a random vector with given marginal distributions and given product-moment or rank correlation matrix. However, this method has been shown to fail to work for some feasible correlation matrices. (A matrix is feasible if there exists a random vector with the given marginal distributions and the matrix as the correlation matrix.) We investigate how this feasibility problem behaves as the dimension of the random vector is increased and find the problem to become acute rapidly. We also find that a modified NORTA procedure, augmented by a semidefinite program (SDP) that aims to generate a correlation matrix "close" to the desired one, performs well with increasing dimension.

[1]  R. M. Cooke,et al.  CONDITIONAL , PARTIAL AND RANK CORRELATION FOR THE ELLIPTICAL COPULA ; DEPENDENCE MODELLING IN UNCERTAINTY ANALYSIS , 2001 .

[2]  R. Saigal,et al.  Handbook of semidefinite programming : theory, algorithms, and applications , 2000 .

[3]  Shane G. Henderson,et al.  Chessboard Distributions and Random Vectors with Specified Marginals and Covariance Matrix , 2002, Oper. Res..

[4]  Henry Wolkowicz,et al.  Handbook of Semidefinite Programming , 2000 .

[5]  Shane G. Henderson,et al.  Simulation mathematics and random number generation: mathematics for simulation , 2001, WSC '01.

[6]  W. Kruskal.  Ordinal Measures of Association , 1958 .

[7]  Averill M. Law,et al.  Simulation Modeling and Analysis , 1982 .

[8]  Roger M. Cooke,et al.  Generating "dependent" quasi-random numbers , 2000, 2000 Winter Simulation Conference Proceedings (Cat. No.00CH37165).

[9]  C. Richard Johnson,et al.  Matrix Completion Problems: A Survey , 1990 .

[10]  Shane G. Henderson,et al.  Chessboard distributions , 2001, Proceeding of the 2001 Winter Simulation Conference (Cat. No.01CH37304).

[11]  Joseph L. Hammond,et al.  Generation of Pseudorandom Numbers with Specified Univariate Distributions and Correlation Coefficients , 1975, IEEE Transactions on Systems, Man, and Cybernetics.

[12]  Shane G. Henderson,et al.  Mathematics for simulation , 2000, Proceeding of the 2001 Winter Simulation Conference (Cat. No.01CH37304).

[13]  Philip M. Lurie,et al.  An Approximate Method for Sampling Correlated Random Variables From Partially-Specified Distributions , 1998 .

[14]  R. Iman,et al.  A distribution-free approach to inducing rank correlation among input variables , 1982 .

[15]  W. Whitt.  Bivariate Distributions with Given Marginals , 1976 .

[16]  R. Clemen,et al.  Correlations and Copulas for Decision and Risk Analysis , 1999 .