Estimnet



Code      Publications     About us


Estimnet is a program for the statistical analysis of large network data. It computes Maximum Likelihood estimates of parameters of Exponential Random Graph Models.

Estimnet implements recently developed Auxiliary Parameter Markov Chain Monte Carlo method and very recently developed Equilibrium Expectation algorithm.
The developers are Maksym Byshkin and Alex Stivala. Its current version may be downloaded from here and from https://github.com/stivalaa/EstimNetDirected

Estimnet is designed for big data. It is applicable to networks on approximately 1,000 to 100,000 nodes. Please refer to Statnet package or PNet program if you want to study smaller networks, or to Bergm package for Bayesian parameter estimation.


Publications

Alexander Borisenko, Maksym Byshkin, Alessandro Lomi, A Simple Algorithm for Scalable Monte Carlo Inference arXiv preprint arXiv:1901.00533 (2019)

Maksym Byshkin, Alex Stivala, Antonietta Mira, Garry Robins, Alessandro Lomi, Fast Maximum Likelihood estimation via Equilibrium Expectation for Large Network Data, Scientific Reports 8:11509 (2018)  [preprint]

Byshkin M, Stivala A, Mira A, Krause R, Robins G, Lomi A, Auxiliary Parameter MCMC for Exponential Random Graph ModelsJournal of Statistical Physics 165: 740-754 (2016) [preprint]

Conference presentations


Fast maximum likelihood estimation via equilibrium expectation for large network data, 13th International Conference in Monte Carlo & Quasi-Monte Carlo Methods in Scientific Computing, July 1-6, 2018, Rennes, France

Maximum Likelihood estimation via Equilibrium Expectation for large network data, Sunbelt
INSNA Conference, June 26 - July 1, 2018, Utrecht, Netherlands

Fast maximum likelihood estimation via equilibrium expectation for large network data, Swiss Numeric Day, April 20th,  2018, Z├╝rich, Switzerland

Efficient Markov chain Monte Carlo Estimation of Exponential Random Graph Models, Second Australian Social Network Analysis Conference, November 28 - 29, 2017, Sydney, Australia

Efficient Markov chain Monte Carlo Estimation of Exponential Random Graph Models, Third European Conference on Social Networks, September 26 - 29, 2017, Mainz, Germany

Efficient Markov chain Monte Carlo Estimation of Exponential Random Graph Models, International Conference on Computational Social Science, July 10 - 13, 2017, Cologne, Germany

Efficient Markov chain Monte Carlo estimation of exponential-family random graph models, International Conference on Monte Carlo Methods and Applications, July 3 - 7, 2017, Montreal, Canada

Scalable MCMC algorithm for the accurate estimation of Exponential Random Graph Models, PASC17, June 26 - 28, 2017, Lugano, Switzerland

Fast maximum likelihood estimation via MCMC equilibrium expectation for the statistical analysis of large networks, Cambridge Networks Day, 13th June 2017, Cambridge, UK

Efficient MCMC Estimation for Exponential Random Graph Models, Sunbelt INSNA Conference, May 30 - June 4, 2017, Beijing, China

Efficient MCMC estimation of structural features of social and other networks, Complex Networks, March 20 - 24, 2017, Dubrovnik, Croatia