This volume introduces the basic concepts of Exponential Random Graph Modeling (ERGM), gives examples of why it is used, and shows the reader how to conduct basic ERGM analyses in their own research. This book fills that gap, by using examples from public health, and walking the reader through the process of ERGM model-building using R statistical software and the statnet package. stream Although it was developed to handle the inherent non-independence of network data, the results of ERGM are interpreted in similar ways to logistic regression, making this a very useful method for examining social systems. �S�ܨ=ر+�}���T�i�I�1̣�p�����u�qEP�`� _��[}�]���itW �����v�92�m@ 3�P`�! We compare these models theoretically, via simulation, and through a real-data example in order to assess their relative strengths and weaknesses. An introduction to exponential random graph (p*) models for social networks Depending on the application, we may consider simple,loopy,multiple-edged, weighted or directed graphs. 1. Thousand Oaks, CA 91320 ERGM is a statistical approach to modeling social network structure that goes beyond the descriptive methods conventionally used in social network analysis. Exponential Family and Random Graph Models (ERGMS) Shengming Luo 3 October 2016 1 Overview De nitions { Graph modeling { Examples: Erd os-Renyi, p 1, 2-star, triangle Properties { Edge prediction { Moments Estimation { MLE equation { Stochastic approximation { MCMCMLE 1.1 De nitions De nition 1.1. If you have not reset your password since 2017, please use the 'forgot password' link below to reset your password and access your SAGE online account. You are in: North America Available Formats . %���� If your library doesn’t have access, ask your librarian to start a trial. In vari-ous applied fields including bioinformatics, speech processing, image processing and control theory, statistical models have long been for-mulated in terms of graphs, and algorithms for computing basic statis- Variational approximations for the exponential random graph model Angelo Mele, Johns Hopkins University Lingjiong Zhu, University of Minnesota We study a model of sequential network formation that converges to the exponential random graph model (ERGM). Extensions of the Basic Model for Directed Networks and Using Dyadic Attributes as Predictors, Appendix B: Modifying R-ergm Model Summary Procedure Using Fix(), Political Science & International Relations, Research Methods, Statistics & Evaluation, Quantitative Applications in the Social Sciences, http://ed.gov/policy/highered/leg/hea08/index.html, CCPA – Do Not Sell My Personal Information. See what’s new to this edition by selecting the Features tab on this page. Recent advances in statistical software have helped make ERGM accessible to social scientists, but a concise guide to using ERGM has been lacking. xڍZY��6~��臭Jw��F�n�%��q�]�L��c�ݜmIl�ÿ~�1�+WY$H� ����D�|�'V��8�Wq{~����l#/J�U*V�Z�����'ϯ���+!�,����f��K�t�������s}����X�٫���m("�ϒ�iD�~���:ot�iS��f+�t��N�F�k� ����lcKm7���JW(*�����'S��ӎM��SQ��䧂��9�e�U����b�|�Lz�I�g�Q����Q���*��V �$�g��獌֦p��dN����Z�[մ�{X����Eg����>~����E�$G�~�H��\�nnx-��Uk���4v$���?��l ź���s�H֟� O�I}��^���4��rc�A�yG {S��G�R�3�-xj5�P�6yC�8l(n!�3�e��G�����Wx�m����B�?�R���ە �sx�X�-}?�殦b�a(3�s��k�}��JfS��K�E�"�vyݝ��B�7ɺ��Է�v� I���;H�!�z�:�oy�`�n�S՗�t$U�+̡E����O�N��7�8#Rk*E%��j՜M�pǽ�N}�����۩��v-��T�rb�6�d�\�D�� _g%T�5}��0�ճ��Qv����|����̧S��TH#H~]�����F !�N-��D=�v*�- SAGE :9��������-V?S?6��y�4*��j�����w^o��}�J�������ď�$~�NB�Sj�S�ms���eMG9OtK"�N�gXѡ����@ż� O�u_�TC��D5��s�.���f�;��k�TyFT�c��q;m��mG�d�OE5��KF�Y�B��\U�CX�ek�S3�v�W+��3}}hL�l,�׼E��m�f=K��~ ��Jw. The temporal exponential random graph model (TERGM) and the stochastic actor-oriented model (SAOM, e.g., SIENA) are popular models for longitudinal net-work analysis. The Promise and Challenge of Network Approaches, 3. For assistance with your order: Please email us at textsales@sagepub.com or connect with your SAGE representative. >> Click the "Preview" tab above to download: This title is also available on SAGE Knowledge, the ultimate social sciences online library. Should you need additional information or have questions regarding the HEOA information provided for this title, including what is new to this edition, please email sageheoa@sagepub.com. De nition: Let G n be the set of all graphs on n vertices. %PDF-1.5 Hello, would you like to continue browsing the SAGE website? 2455 Teller Road Exponential Random Graph Models • Exponential family distribution over networks θ Observed network adjacency matrix Binary indicator for edge (i,j) Features • Properties of the network considered important • Independence assumptions Parameters to be learned Normalizing constant: y ij p(Y = y|θ)= 1 Z eθT φ(y) φ(y) y! Exponential random graph models are a family of probability distributions on graphs. Graphical models bring together graph theory and probability theory in a powerful formalism for multivariate statistical modeling. Building a Useful Exponential Random Graph Model, 4. Exponential Random Graph Models MRQAP 15/108 15/108 Density = 0.21, average degree = 7.4. Learn more about the QASS series here. Please include your name, contact information, and the name of the title for which you would like more information. 3 0 obj << For information on the HEOA, please go to http://ed.gov/policy/highered/leg/hea08/index.html. An Introduction to Exponential Random Graph Modeling is a part of SAGE’s Quantitative Applications in the Social Sciences (QASS) series, which has helped countless students, instructors, and researchers learn cutting-edge quantitative techniques. /Length 3702 /Filter /FlateDecode In-degrees vary from 2 to 16, out-degrees from 0 to 21. www.sagepub.com. Change location, December 2013 | 136 pages | SAGE Publications, Inc.