Motivated by the increasing abundance of data describing real-world networks that exhibit dynamical features, we propose an extension of the Exponential RandomGraph Models (ERGMs) that accommodates the time variation of its parameters. Inspired by the fast growing literature on Dynamic Conditional Score-driven models each parameter evolves according to an updating rule driven by the score of the ERGM distribution. We demonstrate the flexibility of the score-driven ERGMs (SD-ERGMs), both as data generating processes and as filters, and we show the advantages of the dynamic version with respect to the static one. We discuss two applications to time-varying networks from financial and political systems. First, we consider the prediction of future links in the Italian inter-bank credit network. Second, we show that the SD-ERGM allows to discriminate between static or time-varying parameters when used to model the dynamics of the US congress co-voting network.
翻译:我们提议扩展能容纳其参数时间变化的 " 指数随机格格 " 模型(ERGMs),受动态条件计分模型快速增长的文献的启发,每个参数根据ERGM分布得分驱动的更新规则演变。我们展示了作为数据生成过程和过滤器的以分数驱动的ERGMs(SD-ERGMs)的灵活性,并展示了动态版本对静态版本的优势。我们讨论了金融和政治系统对时间变化网络的两种应用。首先,我们考虑了意大利银行间信用网络未来连接的预测。第二,我们表明SD-ERGM允许在用来模拟美国国会联合投票网络动态时区分静态或时间变化参数。