Figure 17

Distributions \(\operatorname{Pr}(k_{i})\) (black ‘∘’) and \(\operatorname{Pr}(k_{i}^{(s)})\) (red ‘□’) for \(M=10^{3}\) Monte Carlo realizations. The distributions are similar but \(\operatorname{Pr}(k_{i}^{(s)})\) systematically deviates to become more skewed as \(k_{i},k_{i}^{(s)}\) increase, signaling that fixing the overall flow through a node and allowing the degree to emerge as a consequence of this leads to larger degree than in observation