scientific article; zbMATH DE number 1163777
From MaRDI portal
Publication:4395385
It will be shown that the maximum entropy distribution with given marginals and normal score correlations can be simply constructed. Sampling from this distribution is very easy too. In contrast, when the dependence is specified through rank correlations, the solution is much more complicated, and an elaborate procedure is required to sample from an approximation of the maximum entropy distribution. Thus, if the choice of measure of dependence is immaterial in the problem definition, normal score correlations are to be preferred. For the entire collection see [Zbl 0885.00054].
Recommendations
Cited in
(10)- On the recovery of joint distributions from limited information
- Maximum entropy distribution of order statistics with given marginals
- A family of composite discrete bivariate distributions with uniform marginals for simulating realistic and challenging optimization-problem instances
- Integrating correlated Bayesian networks using maximum entropy
- Minimally informative distributions with given rank correlation for use in uncertainty analysis
- Maximum entropy in the mean methods in propensity score matching for interval and noisy data
- Using second-order probabilities to make maximum entropy approach to copulas more reasonable
- scientific article; zbMATH DE number 7660126 (Why is no real title available?)
- A proper scoring rule for minimum information bivariate copulas
- Minimum information dependence modeling
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4395385)
