Inference for Network Regression Models with Community Structure

Mengjie Pan, Tyler Mccormick, Bailey Fosdick
Proceedings of the 38th International Conference on Machine Learning, PMLR 139:8349-8358, 2021.

Abstract

Network regression models, where the outcome comprises the valued edge in a network and the predictors are actor or dyad-level covariates, are used extensively in the social and biological sciences. Valid inference relies on accurately modeling the residual dependencies among the relations. Frequently homogeneity assumptions are placed on the errors which are commonly incorrect and ignore critical natural clustering of the actors. In this work, we present a novel regression modeling framework that models the errors as resulting from a community-based dependence structure and exploits the subsequent exchangeability properties of the error distribution to obtain parsimonious standard errors for regression parameters.

Cite this Paper

BibTeX
@InProceedings{pmlr-v139-pan21a, title = {Inference for Network Regression Models with Community Structure}, author = {Pan, Mengjie and Mccormick, Tyler and Fosdick, Bailey}, booktitle = {Proceedings of the 38th International Conference on Machine Learning}, pages = {8349--8358}, year = {2021}, editor = {Meila, Marina and Zhang, Tong}, volume = {139}, series = {Proceedings of Machine Learning Research}, month = {18--24 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v139/pan21a/pan21a.pdf}, url = {https://proceedings.mlr.press/v139/pan21a.html}, abstract = {Network regression models, where the outcome comprises the valued edge in a network and the predictors are actor or dyad-level covariates, are used extensively in the social and biological sciences. Valid inference relies on accurately modeling the residual dependencies among the relations. Frequently homogeneity assumptions are placed on the errors which are commonly incorrect and ignore critical natural clustering of the actors. In this work, we present a novel regression modeling framework that models the errors as resulting from a community-based dependence structure and exploits the subsequent exchangeability properties of the error distribution to obtain parsimonious standard errors for regression parameters.} }
Endnote
%0 Conference Paper %T Inference for Network Regression Models with Community Structure %A Mengjie Pan %A Tyler Mccormick %A Bailey Fosdick %B Proceedings of the 38th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2021 %E Marina Meila %E Tong Zhang %F pmlr-v139-pan21a %I PMLR %P 8349--8358 %U https://proceedings.mlr.press/v139/pan21a.html %V 139 %X Network regression models, where the outcome comprises the valued edge in a network and the predictors are actor or dyad-level covariates, are used extensively in the social and biological sciences. Valid inference relies on accurately modeling the residual dependencies among the relations. Frequently homogeneity assumptions are placed on the errors which are commonly incorrect and ignore critical natural clustering of the actors. In this work, we present a novel regression modeling framework that models the errors as resulting from a community-based dependence structure and exploits the subsequent exchangeability properties of the error distribution to obtain parsimonious standard errors for regression parameters.
APA
Pan, M., Mccormick, T. & Fosdick, B.. (2021). Inference for Network Regression Models with Community Structure. Proceedings of the 38th International Conference on Machine Learning, in Proceedings of Machine Learning Research 139:8349-8358 Available from https://proceedings.mlr.press/v139/pan21a.html.

Related Material