This folder contains R scripts related to the publication of Signorelli, M., Wit, E. C. (2020). Model-based clustering for populations of networks. Statistical Modelling, 20 (1). You can read the paper (with open access) here: https://journals.sagepub.com/doi/full/10.1177/1471082X19871128
This repository contains the data and code to reproduce the simulations presented in Signorelli and Wit (2020).
The code to run each simulation is divided into scripts sequentially numbered:
- script 1 generates the data
- script 2 runs the Expectation-Maximization algorithm for all subcases and repetitions considered (this may in some cases be time consuming, so we provide an implementation that allows to easily perform parallel computing)
- script 3 further processes the output of script 2 to generate the results presented in the paper (with the exception of simulation J, where first model selection is performed in script 3 and then the results are produced in script 4).
Note that scripts 1A, 1B, ..., 2A, 2B, ... require to source some scripts whose name start with 0-; these scripts can be found in the folder "functions to source".