Measuring the Graph Concordance of Locally Dependent Observations
提出一种衡量网络相邻节点结果相关性强于非相邻节点的图一致性指标，并开发了基于置换的置信区间，用于推断网络在解释横截面依赖中的实证作用。
This paper introduces a simple measure of a concordance pattern among observed outcomes along a network, that is, the pattern in which adjacent outcomes tend to be more strongly correlated than nonadjacent outcomes. The graph concordance measure can be generally used to quantify the empirical relevance of a network in explaining cross-sectional dependence of the outcomes, and as shown in the paper, it can also be used to quantify the extent of homophily under certain conditions. When one observes a single large network, it is nontrivial to make inferences about the concordance pattern. Assuming a dependency graph, this paper develops a permutation-based confidence interval for the graph concordance measure. The confidence interval is valid in finite samples when the outcomes are exchangeable, and under the dependency graph, an assumption together with other regularity conditions, is shown to exhibit asymptotic validity. Monte Carlo simulation results show that the validity of the permutation method is more robust than the asymptotic method to various graph configurations.