Statistical analysis of network data is an active field of study, in which researchers inves- tigate graph-theoretic concepts and various probability models that explain the behaviour of real networks. This thesis attempts to combine two of these concepts: an exponential random graph and a centrality index. Exponential random graphs comprise the most useful class of probability models for network data. These models often require the assumption of a complex dependence structure, which creates certain difficulties in the estimation of unknown model parameters. However, in the context of dynamic networks the exponential random graph model provides the opportunity to incorporate a complex network structure such as centrality without the usual drawbacks associated with parameter estimation. The thesis employs this idea by proposing probability models that are equivalent to the logistic regression models and that can be used to explain behaviour of both static and dynamic networks.

Statistical analysis of network data is an active field of study, in which researchers inves- tigate graph-theoretic concepts and various probability models that explain the behaviour of real networks. This thesis attempts to combine two of these concepts: an exponential random graph and a centrality index. Exponential random graphs comprise the most useful class of probability models for network data. These models often require the assumption of a complex dependence structure, which creates certain difficulties in the estimation of unknown model parameters. However, in the context of dynamic networks the exponential random graph model provides the opportunity to incorporate a complex network structure such as centrality without the usual drawbacks associated with parameter estimation. The thesis employs this idea by proposing probability models that are equivalent to the logistic regression models and that can be used to explain behaviour of both static and dynamic networks.

<p>This thesis presents Bayesian solutions to inference problems for three types of social network data structures: a single observation of a social network, repeated observations on the same social network, and repeated observations on a social network developing through time.</p><p>A social network is conceived as being a structure consisting of actors and their social interaction with each other. A common conceptualisation of social networks is to let the actors be represented by nodes in a graph with edges between pairs of nodes that are relationally tied to each other according to some definition. Statistical analysis of social networks is to a large extent concerned with modelling of these relational ties, which lends itself to empirical evaluation.</p><p>The first paper deals with a family of statistical models for social networks called exponential random graphs that takes various structural features of the network into account. In general, the likelihood functions of exponential random graphs are only known up to a constant of proportionality. A procedure for performing Bayesian inference using Markov chain Monte Carlo (MCMC) methods is presented. The algorithm consists of two basic steps, one in which an ordinary Metropolis-Hastings up-dating step is used, and another in which an importance sampling scheme is used to calculate the acceptance probability of the Metropolis-Hastings step.</p><p>In paper number two a method for modelling reports given by actors (or other informants) on their social interaction with others is investigated in a Bayesian framework. The model contains two basic ingredients: the unknown network structure and functions that link this unknown network structure to the reports given by the actors. These functions take the form of probit link functions. An intrinsic problem is that the model is not identified, meaning that there are combinations of values on the unknown structure and the parameters in the probit link functions that are observationally equivalent. Instead of using restrictions for achieving identification, it is proposed that the different observationally equivalent combinations of parameters and unknown structure be investigated a posteriori. Estimation of parameters is carried out using Gibbs sampling with a switching devise that enables transitions between posterior modal regions. The main goal of the procedures is to provide tools for comparisons of different model specifications.</p><p>Papers 3 and 4, propose Bayesian methods for longitudinal social networks. The premise of the models investigated is that overall change in social networks occurs as a consequence of sequences of incremental changes. Models for the evolution of social networks using continuos-time Markov chains are meant to capture these dynamics. Paper 3 presents an MCMC algorithm for exploring the posteriors of parameters for such Markov chains. More specifically, the unobserved evolution of the network in-between observations is explicitly modelled thereby avoiding the need to deal with explicit formulas for the transition probabilities. This enables likelihood based parameter inference in a wider class of network evolution models than has been available before. Paper 4 builds on the proposed inference procedure of Paper 3 and demonstrates how to perform model selection for a class of network evolution models.</p>

This thesis presents Bayesian solutions to inference problems for three types of social network data structures: a single observation of a social network, repeated observations on the same social network, and repeated observations on a social network developing through time. A social network is conceived as being a structure consisting of actors and their social interaction with each other. A common conceptualisation of social networks is to let the actors be represented by nodes in a graph with edges between pairs of nodes that are relationally tied to each other according to some definition. Statistical analysis of social networks is to a large extent concerned with modelling of these relational ties, which lends itself to empirical evaluation. The first paper deals with a family of statistical models for social networks called exponential random graphs that takes various structural features of the network into account. In general, the likelihood functions of exponential random graphs are only known up to a constant of proportionality. A procedure for performing Bayesian inference using Markov chain Monte Carlo (MCMC) methods is presented. The algorithm consists of two basic steps, one in which an ordinary Metropolis-Hastings up-dating step is used, and another in which an importance sampling scheme is used to calculate the acceptance probability of the Metropolis-Hastings step. In paper number two a method for modelling reports given by actors (or other informants) on their social interaction with others is investigated in a Bayesian framework. The model contains two basic ingredients: the unknown network structure and functions that link this unknown network structure to the reports given by the actors. These functions take the form of probit link functions. An intrinsic problem is that the model is not identified, meaning that there are combinations of values on the unknown structure and the parameters in the probit link functions that are observationally equivalent. Instead of using restrictions for achieving identification, it is proposed that the different observationally equivalent combinations of parameters and unknown structure be investigated a posteriori. Estimation of parameters is carried out using Gibbs sampling with a switching devise that enables transitions between posterior modal regions. The main goal of the procedures is to provide tools for comparisons of different model specifications. Papers 3 and 4, propose Bayesian methods for longitudinal social networks. The premise of the models investigated is that overall change in social networks occurs as a consequence of sequences of incremental changes. Models for the evolution of social networks using continuos-time Markov chains are meant to capture these dynamics. Paper 3 presents an MCMC algorithm for exploring the posteriors of parameters for such Markov chains. More specifically, the unobserved evolution of the network in-between observations is explicitly modelled thereby avoiding the need to deal with explicit formulas for the transition probabilities. This enables likelihood based parameter inference in a wider class of network evolution models than has been available before. Paper 4 builds on the proposed inference procedure of Paper 3 and demonstrates how to perform model selection for a class of network evolution models.

Η παρούσα διπλωματική εργασία βρίσκεται στον ευρύτερο χώρο της μαθηματικής στατιστικής θεωρίας των γραφημάτων. Κύριος στόχος μας, όπως αναφέρει και ο τίτλος, είναι η μοντελοποίηση γραφημάτων, με απώτερο σκοπό την προσαρμογή, προσομοίωση και διάγνωση αυτών μέσω μοντέλων εκθετικών τυχαίων γραφημάτων. Το πρώτο κεφάλαιο δίνει μια συνοπτική παρουσίαση της διατύπωσης του προβλήματος και της θεωρίας των μοντέλων των εκθετικών τυχαίων γραφημάτων. Η βασική ιδέα είναι να θεωρήσουμε ως τυχαίες μεταβλητές τους δυνατούς δεσμούς μεταξύ των κόμβων ενός δοθέντος γραφήματος. Η γενική μορφή ενός μοντέλου εκθετικά τυχαίου γραφήματος καθορίζεται από κάποιες υποθέσεις σχετικές με τις εξαρτήσεις μεταξύ αυτών των τυχαίων μεταβλητών. Παρουσιάζουμε κάποιες διαφορετικές υποθέσεις εξάρτησης και τα αντίστοιχα μοντέλα, όπως τα γραφημάτα Bernoulli, τα δυαδικώς - ανεξάρτητα και τα τυχαία γραφήματα Markov. Επίσης, εξετάζουμε την ενσωμάτωση των χαρακτηριστικών, που μπορούν να έχουν οι κόμβοι, σε μοντέλα κοινωνικής επιλογής, δηλαδή, σε περιπτώσεις που οι συνδέσεις του γραφήματος μπορούν να προβλέψουν τα χαρακτηριστικά των κόμβων. Συνοψίζουμε κάποιες καινούργιες υποθέσεις εξάρτησης, που είναι πολυπλοκότερες των πρώτων τέτοιων υποθέσεων της σχετικής βιβλιογραφίας. Συζητούμε τις διαδικασίες της στατιστικής εκτίμησης, συμπεριλαμβανομένων των νέων μεθόδων για την εκτίμηση της μέγιστης πιθανοφάνειας Monte Carlo. Τέλος, παρουσιάζουμε τις νέες προδιαγραφές για μοντέλα εκθετικών τυχαίων γραφημάτων, που έχουν προτείνει οι Snijders et al., οι οποίες βελτιώνουν σημαντικά τα αποτελέσματα της προσαρμογής εμπειρικών δεδομένων για εκθετικά μοντέλα ομοιογενών τυχαίων γραφημάτων Markov. Επιπλέον, οι νέες αυτές προδιαγραφές μας βοηθούν να αποφύγουμε το πρόβλημα του σχεδόν-εκφυλισμού, που συχνά παρεμβάλλεται στη διαδικασία της προσαρμογής μοντέλων εκθετικών τυχαίων γραφημάτων Markov, ιδιαίτερα όταν αυτά προέρχονται από εμπειρικά δεδομένα, που έχουν υψηλό βαθμό μεταβατικότητας. Η μελέτη μιας τέτοιας νέας στατιστικής με υψηλότερης τάξης μεταβατικότητα επιτρέπει την εκτίμηση των παραμέτρων των μοντέλων των εκθετικών γραφημάτων σε πολλές (αλλά όχι όλες) περιπτώσεις, στις οποίες διαφορετικά θα ήταν αδύνατο να εκτιμηθούν οι παράμετροι των μοντέλων των ομοιογενών γραφημάτων Markov. Στο δεύτερο, τρίτο και τέταρτο κεφάλαιο της εργασίας εφαρμόζουμε τις παραπάνω μεθόδους, αντιστοίχως, για τρείς αναλύσεις εμπειρικών δεδομένων: το δίκτυο Florentine, το δίκτυο Faux Magnolia High και τα δίκτυα IPRED και SWPAT. Σε αυτά τα κεφάλαια, παρουσιάζουμε τις διαδικασίες της προσαρμογής, προσομοίωσης και διάγνωσης με παράθεση των αντίστοιχων εντολών, χρησιμοποιώντας τα πακέτα statnet - ermg και sna, τα οποία δουλεύουν στο περιβάλλον του πακέτου ελεύθερου λογισμικού R. Τέλος, στο παράρτημα της εργασίας δίνουμε μια σύντομη εισαγωγή στο περιβάλλον R και σε κάποιες γενικές εντολές αυτού. / This specific project has to do with mathematical statistical graph theory. Our main target is to fit, simulate and diagnose models through exponential random graph models. In the first chapter we give a short presentation of the problem and the theory of exponential random graph models. The main idea is to consider each tie of a given network (graph) as a random variable. The general form of an exponential random graph model is defined from some relative assumptions that have to do with the dependence between those random variables. We present some different dependence assumptions and the corresponding models, such as Bernoulli graphs, dyadic-independent and Markov random graphs. We also examine the incorporation of the characteristics that a node may have in social networks. We also discuss the process of statistical estimation, including three new methods for the estimation of Monte Carlo maximum likelihood. Finally, we present new specifications for exponential random graph models, which Snijders et al. have proposed. These new specifications allow us to avoid the problem of degeneration. In the second, third and fourth chapter we apply the above methods in order to analyze Florentine network data, Faux Magnolia High data and IPred And Swpat data. In those chapters, we present the procedures of fit, simulate and diagnose exponential random graph models displaying the corresponding commands of statnet-ergm and sna packages that work in R. Finally we give a short introduction to R and to some relative commands.