TY - JOUR
T1 - Network inference from temporally dependent grouped observations
AU - Zhao, Yunpeng
N1 - Funding Information:
This research was supported by NSF Grant DMS 1840203. We thank Dr. Richard Wrangham for sharing the research results of the Kibale Chimpanzee Project. We thank Dr. Charles Weko for preparing and cleaning the data sets.
Funding Information:
This research was supported by NSF Grant DMS 1840203 . We thank Dr. Richard Wrangham for sharing the research results of the Kibale Chimpanzee Project. We thank Dr. Charles Weko for preparing and cleaning the data sets.
Publisher Copyright:
© 2022 Elsevier B.V.
PY - 2022/7
Y1 - 2022/7
N2 - In social network analysis, the observed data usually reflect certain social behaviors, such as the formation of groups, rather than an explicit network structure. Zhao and Weko proposed a model-based approach called the hub model to infer implicit networks from grouped observations (Zhao and Weko, 2019). The hub model assumes independence between groups, which sometimes is not valid in practice. The hub model is generalized into the case of grouped observations with temporal dependence. As in the hub model, the group at each time point is gathered under one leader in the new model. Unlike in the hub model, the group leaders are not sampled independently but follow a Markov chain, and other members in adjacent groups can also be correlated. An expectation-maximization (EM) algorithm is developed for this model and a polynomial-time algorithm is proposed for the E-step. The performance of the new model is evaluated under different simulation settings. The proposed model is applied to a data set of the Kibale Chimpanzee Project.
AB - In social network analysis, the observed data usually reflect certain social behaviors, such as the formation of groups, rather than an explicit network structure. Zhao and Weko proposed a model-based approach called the hub model to infer implicit networks from grouped observations (Zhao and Weko, 2019). The hub model assumes independence between groups, which sometimes is not valid in practice. The hub model is generalized into the case of grouped observations with temporal dependence. As in the hub model, the group at each time point is gathered under one leader in the new model. Unlike in the hub model, the group leaders are not sampled independently but follow a Markov chain, and other members in adjacent groups can also be correlated. An expectation-maximization (EM) algorithm is developed for this model and a polynomial-time algorithm is proposed for the E-step. The performance of the new model is evaluated under different simulation settings. The proposed model is applied to a data set of the Kibale Chimpanzee Project.
KW - Forward-backward algorithm
KW - Grouping behavior
KW - Social networks
UR - http://www.scopus.com/inward/record.url?scp=85126543252&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85126543252&partnerID=8YFLogxK
U2 - 10.1016/j.csda.2022.107470
DO - 10.1016/j.csda.2022.107470
M3 - Article
AN - SCOPUS:85126543252
SN - 0167-9473
VL - 171
JO - Computational Statistics and Data Analysis
JF - Computational Statistics and Data Analysis
M1 - 107470
ER -