Aiming to understand real-world hierarchical networks whose degree distributions are neither power law nor exponential, we construct a hybrid clique network that includes both homogeneous and inhomogeneous parts, and introduce an inhomogeneity parameter to tune the ratio between the homogeneous...

We study the structural properties of Russian stock market by means of analysis of the corresponding market graph. For this graph we found the distribution of correlations, edge density, maximum cliques and maximum independent sets. We also study the evolution of the structural properties of the...

Given a multinomial decomposable graphical model, we identify several alternative parametrizations; in particular we consider conditional probabilities of clique-residuals given separators, as well as generalized log-odds-ratios. For each such parametrization, we construct the corresponding...