Multivariate Distributions with Fixed Marginals and Correlations
Consider the problem of drawing random variates (X 1, …, X n ) from a distribution where the marginal of each X i is specified, as well as the correlation between every pair X i and X j . For given marginals, the Fréchet-Hoeffding bounds put a lower and upper bound on the correlation between X i and X j . Any achievable correlation between X i and X j is a convex combination of these bounds. We call the value λ(X i , X j ) ∈ [0, 1] of this convex combination the convexity parameter of (X i , X j ) with λ(X i , X j ) = 1 corresponding to the upper bound and maximal correlation. For given marginal distributions functions F 1, …, F n of (X 1, …, X n ), we show that λ(X i , X j ) = λ ij if and only if there exist symmetric Bernoulli random variables (B 1, …, B n ) (that is {0, 1} random variables with mean ½) such that λ(B i , B j ) = λ ij . In addition, we characterize completely the set of convexity parameters for symmetric Bernoulli marginals in two, three, and four dimensions.