Introduction to
Stochastic Actor-oriented Models
Application of SAOMs
András VörösEntrepreneurship and Innovation Assignment
Department of Social Statistics, University of Manchester
SOST71032 Social Network Analysis
* Thanks to Christoph Stadtfeld for the nice figure in this presentation.
How to apply SAOMs in practice?
 SAOMs are implemented in the RSiena software
 In R, you need the RSiena package to run the model
 You need data in appropriate format
 Dependent networks: squared adjacency matrix per obs.
(n*n*t), binary tie values, one- or two mode, directed or undir.
 Dependent behaviors: matrix of values per obs. (n*t)
 Constant actor covariates: vector of values (n*1)
 Changing actor covariates: matrix of values per period (n*(t-1))
 Constant dyadic covariates: squared adjacency matrix per
period (n*n*(t-1)), arbitrary tie values
 Changing dyadic covariates:
 And need to use RSiena’s own object classes
 (More specific details will come up later)
How to apply SAOMs in practice?
Please continue with the next topic.