论文标题
有效的签名网络中签名的$β$模型的有效估计和推断
Efficient estimation and inference for the signed $β$-model in directed signed networks
论文作者
论文摘要
本文提出了一种新颖的签名$β$ - 模型，用于定向签名的网络，该网络经常在应用域中遇到，但在文献中很大程度上被忽略了。提出的签名$β$ - 模型分解了一个有向签名的网络，因为两个无符号网络的差异，每个节点嵌入了两个潜在因素，用于现场和现场。负边的存在导致非cove log-likelihienhiehoens，并且开发了一步估计算法以促进参数估计，这在理论上和计算上都是有效的。我们还为签名的$β$ - 模型下的成对和多个节点比较制定了推论程序，这填补了缺乏节点排名的不确定性定量的空白。为置信区间的覆盖范围以及多个节点比较的错误发现率（FDR）控制建立了理论结果。还通过对合成网络和现实生活中的广泛的数值实验来检查签名的$β$模型的有限样品性能。
This paper proposes a novel signed $β$-model for directed signed network, which is frequently encountered in application domains but largely neglected in literature. The proposed signed $β$-model decomposes a directed signed network as the difference of two unsigned networks and embeds each node with two latent factors for in-status and out-status. The presence of negative edges leads to a non-concave log-likelihood, and a one-step estimation algorithm is developed to facilitate parameter estimation, which is efficient both theoretically and computationally. We also develop an inferential procedure for pairwise and multiple node comparisons under the signed $β$-model, which fills the void of lacking uncertainty quantification for node ranking. Theoretical results are established for the coverage probability of confidence interval, as well as the false discovery rate (FDR) control for multiple node comparison. The finite sample performance of the signed $β$-model is also examined through extensive numerical experiments on both synthetic and real-life networks.
