Siena: Statistical Modeling of Longitudinal Network Data
- Network panel data
longitudinal data consisting of two or more repeated observations of a network on a given set of nodes.
- Panel waves
the data observed for one given observation moment in a panel study.
- Social actors
individuals, companies, etc., represented by the nodes in the network.
- Stochastic actor-oriented model
a probability model for network dynamics where changes may take place at arbitrary moments in continuous time, and where these changes are regarded as consequences of choices made by the actors.
R package implementing statistical inference according to a stochastic actor-oriented model given network panel data.
model components defining the probabilities of tie changes in the stochastic actor-oriented model.
- Method of Moments
one of the traditional methods in statistics for parameter estimation.
- Dependent variable
Work on the Siena method and software has taken place at the University of Groningen (Department of Sociology) and the University of Oxford (Department of Statistics; Department of Politics and International Relations; Nuffield College). NWO (the Netherlands Organisation for Scientific Research) has supported development of this methodology and software through the project Statistical Methods for the Joint Development of Individual Behavior and Peer Networks (project number 575-28-012, researcher Mark Huisman), the integrated research program The Dynamics of Networks and Behavior (project number 401-01-550, methodological researchers Christian Steglich and Michael Schweinberger), and the ECRP-Eurocores program Models for the Evolution of Networks and Behavior (project number 461-05-690, methodological researcher Christian Steglich). The US National Institutes of Health have provided funding for methodological developments (Johan Koskinen) and software development (Ruth Ripley and Krists Boitmanis) as part of the project Adolescent Peer Social Network Dynamics and Problem Behavior (grant number 1R01HD052887-01A2).
- Lospinoso JA (2012) Statistical models for social network Dynamics. DPhil Thesis, University of Oxford, OxfordGoogle Scholar
- Ripley RM, Snijders TAB, Preciado P (2017) Manual for Siena version 4.0. Technical report, Department of Statistics, University of Oxford, Oxford/Nuffield College. http://www.stats.ox.ac.uk/siena/. Accessed 28 May 2013
- Snijders TAB (1995) Methods for longitudinal social network data: Review and Markov process models. In: Tiit E, Kollo T, Niemi H (eds) New Trends in Probability and Statistics Vol. 3: Multivariate Statistics and Matrices in Statistics. Proceedings of the 5th Tartu Conference, TEV Vilnius, Lithuania, pp 211–227Google Scholar
- Snijders TAB (2001) The statistical evaluation of social network dynamics. In: Sobel ME, Becker MP (eds) Sociological methodology – 2001, vol 31. Basil Blackwell, Boston and London, pp 361–395Google Scholar
- Snijders TAB, Steglich CEG, Schweinberger M (2007) Modeling the co-evolution of networks and behavior. In: van Montfort K, Oud H, Satorra A (eds) Longitudinal models in the behavioral and related sciences. Lawrence Erlbaum, Mahwah, pp 41–71Google Scholar
- van de Bunt GG (1999) Friends by choice; an actor-oriented statistical network model for friendship networks through time. Thesis Publishers, AmsterdamGoogle Scholar
- Wasserman S (1979) A stochastic model for directed graphs with transition rates determined by reciprocity. In: Schuessler KF (ed) Sociological methodology 1980. Jossey-Bass, San FranciscoGoogle Scholar
- In addition to the help pages that are available as for all R packages, there is an extensive manual (Ripley et al. 2017) and a tutorial paper (Snijders et al. 2010b). A textbook about the Siena method and an edited volume with example applications are in preparation. The website http://www.stats.ox.ac.uk/siena/ is actively maintained and contains references to the basic methodology, references to applications, R scripts, example data sets, workshop announcements, and more
- For those who wish to read more about the mathematical and methodological background, a recommended sequence of readings could be Snijders (1996) as an introduction to the idea of stochastic actor-oriented models, Snijders (2001) or Snijders (2005) for the basic definition of the model for one dependent network defined as a changing digraph, and Steglich et al. (2010) for models for the dynamics of networks and behavior, which might be followed by Snijders et al. (2010a) for Maximum Likelihood estimation or Snijders et al. (2013) for models with multiple dependent networks.Google Scholar