2014 tavaszi előadássorozat előadásai

Seminars

2014 Winter – Spring

 

 

7 January, 2014 @ MTA TK

Nikita Krugljakov (Maastricht University): Much Obliged: On the diffusion of prosocial behavior in a simple network setting.

In a lab experiment I study whether the beneficiaries of a generous act themselves act more generously towards disadvantaged third parties. Participants play a three-player version of the trust game. The trustee decides how to allocate the money received from the trustor between the trustor, herself, and a third party. I find that the trustee positively reciprocates towards the trustor and the third party. However, the relationship between trustor’s giving and trustee’s total giving is U-shaped. The trustee favors the third party (the trustor) when the transfer from the trustor is small (large). My findings add to the growing literature on generalized reciprocity. They suggest that the diffusion of prosocial behavior – often referred to as the pay it forward phenomenon – is less likely to be observed in situations in which the benefactors act very generously towards the beneficiaries and the latter can directly reciprocate towards the former.

 

22 January, 2014 @ CEU

Ales Ziberna (University of Ljubljana): (Generalized) blockmodeling: what, why and how

In this presentation I will introduce the concept of blockmodeling and in particular generalized blockmodeling. Blockmodeling is a technique forfinding clusters of units in the network and, at the same time determining ties within and between these clusters. First I will explain on intuitive what blockmodelingis and for what purposes it can be used. This will be illustrated by several examples. Next I will present an overview of different types of or approaches to blockmodeling. In the last part of the presentation I will focus on generalized blockmodeling and in particular to its extensions for valued and multilevel networks.

 

28 January, 2014 @ MTA TK

András Tóth (MTA Center of Political Sciences): After Malaise: Rise of Selective Economic Nationalism and the Allure of the Sweet Melodies of the Grasshopper Song

This article is the third article in a series of articles written on Hungary. In the first article we noted that Hungary is in a state of certain malaise and uncertainty whether the country is on the right development track (Neumann- Tóth, 2009). Three years later we stated that Hungary is suffering from a full-blown malaise (Tóth-Neumann-Hosszú 2012). The intertwining of a political crisis, domestic debt crisis, begun in 2006, and the world wide credit crunch had resulted in a perfect political storm. This storm led to the elections of FIDESZ led by the charismatic Viktor Orbán. Orbán declared that it took place a revolution and to respond to the desire of the people to change, it would build a new Hungarian model instead of the earlier one, which ended in dismal failure. The stated goal of the government was to create a new more indigenous and independent growth model based on strong government and domestic economic actors.

 

5 February, 2014 @ CEU

Simone Righi (MTA Research Center for Educational and Network Studies): Pricing in Social Networks under Limited Information

This paper models the strategy of a monopolist that offers rewards to current clients in order to induce them to activate their social network and convince peers to buy from the company. In presence of heterogeneous search costs and reservation prices, this network-activation reward program may serve to expand the client base through a flow of information from informed to uninformed consumers. The offer of the monopolist affects individual incentives of aware people to share information, determining a minimal degree condition for investment. The optimal unitary reward balances the information spread effect (i.e. more receivers) and the crowding effect (i.e. less individual incentives) of an increase in the number of speakers. The monopolist always finds it profitable to use the bonus. Nevertheless, its introduction has ambiguous effects on the price and profits, depending on the process of spread of information and, in turn, on the network structure

.

25 February, 2014 @ MTA TK

Balázs Vedres (CNS-CEU)

What accounts for creative success when the unit of innovation is a team? In particular, what are the sociological factors that explain why some ensembles are able to meet the challenge of creating a cultural product that is not only inventive but also critically acclaimed? We build on work pointing to structural folding – the network property of a cohesive group whose membership overlaps with that of another cohesive group. To explore the processes whereby structural folding contributes to creative success, we draw on new insights in cultural sociology. We hypothesize that the effects of structural folding on game changing success are especially strong when overlapping groups are cognitively distant. That is, teams are most likely to produce games that stand out and are recognized as outstanding when their cognitively heterogeneous communities have points of intersection. To test our hypothesis about structural folding and cognitive diversity, we study the historical mechanisms of team reassembly in the video game industry. We collected data on 12,422 video games and the career histories of 139,727 video game developers. Because we measure distinctiveness independently from critical acclaim, we can test whether teams with structural folds that span cognitively distant communities are able to develop distinctive products that are, at the same time, recognized as successful in the video gaming field.

 

4 March, 2014 @ MTA TK

János Kertész (CNS-CEU): Value production in a collaborative internet environment: Wikipedia edit wars

Wikipedia is the paradigmatic example of collaboration on the internet and its advantage is, from the research point of view, that every editing or discussion event is perfectly documented and freely downloadable. We have been interested in the emergence and resolution of conflicts. We constructed a measure, which made it possible to identify controversial articles algorithmically. We investigated the statistical properties of such pages in a number of languages and concluded that, except of a few, generally disputed topics, the most controversial pages are culture-dependent. By looking at the temporal evolution of the conflicts we were able to identify three categories: Isolated conflicts, multiple conflicts and permanent wars. We introduced an opinion dynamics model by which we were able to reproduce qualitatively our findings.

 

11 March, 2014 @ CEU

Bence Ságvári, Júlia Koltai (MTA Institute of Sociology): The growth of an online social network: diffusion and methodological challenges

ICT has changed the availability, communication, contacting and interacting habits of people thus it has become a considerable factor in restructuring the society. Therefore, it is of major interest to understand how ICT related social networks emerge, evolve and eventually decline. It is a fortunate situation that there is a large Hungarian OSN, called iWiW, where all these processes can be studied in detail. The aim of the present project is to analyze the iWiW data from the above points of view.

The early period of an OSN is closely related to the widely studied phenomenon of diffusion of innovations. Most of the studies had to come to terms studies of with modest sizes here we have a sample of the order of million users. Due to the availability of some metadata, we will be able to identify and analyze influencing factors in the diffusion process, such as age, gender and geographical socio-economic attributes.

As a part of ongoing interdisciplinary research project our presentation will focus on the methodological dilemmas of using massive online social networking data and we also present some in-depth “historical” analysis on the spread of the iWiW network.

 

18 March, 2014 @ MTA TK

András Vörös (Oxford University/MTA Research Center for Educational and Network Studies): Cluster Analysis of Multiplex Networks: Dening Composite Network Measures

Multiplex network data, information on multiple relations in a given group, provides researchers an opportunity to study social processes in depth, and to answer questions about the interdependence of dierent relational dimensions. Although some multivariate network methods (e.g. ERGM, SIENA) make it possible to jointly analyze multiple network dimensions, modelling becomes impossibly complex when the investigation focuses on more than a few, say more than three or four, network dimensions. In these cases, dimension reduction methods may be applied to obtain a manageable set of variables. Drawing on existing statistical methods and measures, we propose a procedure to reduce the dimensions of multiplex network data measured in multiple groups. We achieve this by clustering the networks using their pairwise similarities and constructing composite network measures as combinations of the networks in each resulting cluster. The procedure is demonstrated on an example on the dimensions of perceptions about peers in Hungarian high-school class- rooms: starting from 22 perception networks we arrive at a 3-group solution which we label as positive traits, negative traits, and social role attributions. Though our procedure does not rely on an explicit statistical model, it presents a useful and mension reduction in multiplex networks. Following such an approach may aid researchers in dening complex network measures and may also provide some theoretical insights into multiplex social mechanisms.

 

25 March, 2014 @ MTA TK

Christoph Stadtfeld (University of Lugano/University of Groningen): Micromotives and Macrobehavior in Social Networks

Many phenomena in social networks can be well explained by individual motivations and choices (micromotives in social networks). For example, trust, friendship, homophily, behavioral imitation and social learning can be described as small-scale processes with just two or a few individuals involved. However, many of these small-scale processes interconnect to large-scale social networks with specific features: Clustering,  segregation and diffusion of knowledge are examples of complex large-scale structures and processes (macrobehavior in social networks) that emerge from small-scale motives and in turn have an influence on the small-scale level. In this talk, I will discuss the complex interplay between micromotives and macrobehavior in social networks. In particular, I will focus on how modern social networks research can provide new insights into this theoretical challenge by combining different methodological approaches.

 

27 March, 2014 @ CNS

Fabrizio Lillo (Universita' di Palermo, Italy / Santa Fe Institute, United States): Systemic price cojumps

Instabilities in the price dynamics of a large number of financial assets are a clear sign of systemic events. By investigating high cap stocks traded in different stock exchanges, we find that there is a large number of multiple cojumps, i.e. minutes in which a sizable number of stocks displays a discontinuity of the price process. By considering the last fifteen years, which have experienced the introduction of High Frequency Trading, we show that the number of jumps has slightly declined, but the number of cojumps, especially when involving a large number of stocks, has significantly increased. This is a clear sign of an increased synchronisation in financial markets. We introduce a one factor model approach where both the factor and the idiosyncratic jump components are described by a Hawkes process. We introduce a robust calibration scheme which is able to distinguish systemic and idiosyncratic jumps. Finally, we study the exogenous (i.e. news driven) and endogenous nature of systemic price cojumps and we find that only approximately one third of systemic cojumps can be associated with a macroeconomic news. (http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2209139)

 

1 April, 2014 @ CEU

Boleslaw Szymanski (Rensselaer Polytechnic Institute, Troy, NY): Spread of Opinions Dynamics in Social Networks

Human behavior is profoundly affected by the influenceability of individuals and their social networks. This talk discusses the dynamics of spread of opinions in such networks using fundamental models for Social Contagion: the binary agreement model (influencing with committed minorities) and threshold model (threshold contact process). In the first one all individuals initially adopt either opinion A or B, and a small fraction of all individuals commits to their opinions. Committed individuals are immune to influence but otherwise follow the prescribed rules for opinion change. We show that the prevailing majority opinion in a population can be rapidly reversed by a small fraction of randomly distributed committed individuals. When committed individuals exist for both opinions, the difference between larger and smaller fractions of them needed for rapid majority conversion decreases as the smaller minority increases. The results are relevant in understanding and influencing the social perceptions of ideas and policies. We used the threshold model to find efficient spreaders, fast heuristic selection strategies, and impact of clustering on system dynamics. We find that even for arbitrarily high value of threshold, there exists a critical initiator fraction beyond which the cascade becomes global. Network structure, in particular clustering, plays a significant role in this scenario. Similarly to the case of single-node or single-clique initiators studied previously, we observe that community structure within the network facilitates opinion spread to a larger extent than a homogeneous random network. Finally, we study the efficacy of different initiator selection strategies on the size of the cascade and the cascade window.

 

8 April, 2014 @ CEU

Juraj Medzihorsky (CEU, Political Science): The mixture index of fit in social network analysis

The π* mixture index of fit is a new measure of model fit and can be applied also in social network analysis.It is easy to interpret, applies to many different models, and unlike conventional measures of fit it rests on assumptions that are always true. The index measures model fit by the smallest fraction of the population that cannot be described perfectly by a distribution belonging to the model, and facilitates new substantive findings through the analysis of the residual component.  It does not depend on sample size and does not require stochastic samples, which makes it particularly appealing to use with population data. 'pistar,' an R package, allows to conveniently obtain the value of the mixture index in various settings.

 

22 April, 2014 @ CEU

Silvia Neamtu (CNS/CEU): Understanding Political-Business Elite Relations in Hungary through Statistical M odels for Networks

In this report, I present the preliminary findings with respect to what explains a particular network of political and business elites in Hungary. Based on a unique sample of such a network, I conduct exploratory network analysis and build statistical models to test certain ideas suggested by the descriptive analysis. I construct three Exponential Random Graph Models (ERGMs), based on alternative theoretical assertions about what generated the network of political-business ties: the social-dependence hypothesis, the cognitive consistency hypothesis, and the homophily hypothesis. The goal of the research is to understand what micro-level (actor-level) processes have generated the network we observe. ERGMs are an appropriate tool for this research problem, because they can be specified to represent multiple mechanisms that underlie the stochastic generation of the static network explored in this paper. They provide a simple and useful tool for understanding what is the probability of observing the network of interest, based on both features of the network, as well as attributes of the actors involved. The partial findings bring support in favor of the social-dependence hypothesis, suggesting that actors in the investigated network establish ties through minimizing their dependence on others and maximizing others’ dependence on them. Further research is suggested that would explore these networks over time, to uncover the complexity of political-business elite interactions. Keywords: Political networks; political-business elite relations; Hungary; ERGMs; local-global network properties

 

6 May, 2014 @ MTA TK

Carl Nordlund (CNS/CEU): The significance of unexpected anomalies: a deviational approach to blockmodeling of valued networks

What relations are signficant in valued networks? In order to apply the standard set of network methods to valued networks, data is typically dichotomized, i.e. where valued relations are binarized using a statistically, theoretically, or arbtrarily determined absolute threshold value. In addition to the significant loss of details this entails, dichotomization implies an implicit assumption of equal relational capacity among actors. Puerto Rican exports to USA are indeed very significant to the former, while only constituting a fraction of total US imports, distinctions that existing approaches fail to capture. This paper proposes a novel approach to role-analysis and blockmodeling of valued networks that is more sensitive to patterns, rather than mere strengths, of ties. Rather than looking at the absolute values of relations, or examining valued ties on a per-actor basis (cf Nordlund 2007), the approach specifies significance in terms of unexpected anomalies. By comparing the distribution of each actor's valued relations to its alters with the macro-level distributions of total in- and outdegrees, significant (1) and non-significant (0) ties are determined simultaneously on a per-actor-to-actor and a per-actor-from-actor basis. This allows for a direct interpretation of the underlying functional anatomy of a valued network that is explicitly concerned with patterns, rather than mere strengths, of its relations. In addition to its applicability for direct blockmodeling, the article also suggests a novel indirect measure of deviational structural equivalence that similarly identifies roleequivalence on the basis of patterns, rather than strengths, of ties. Exemplified with the notesharing data of Ziberna (2007), the EIES messages data, and total commodity trade among EU/EFTA countries as of 2010, both the direct and indirect approach yields results that, it is argued, are far more intuitive than existing approaches. This is particularly evident in the case of the EU/EFTA trade network, where the indirect approach yields partitions that reveals patterns in support of theories of intra-regional trade. Summarizing the approach and the empirical findings, the conclusion also outlines how the suggested deviational approach can be transposed into other metrics at the micro- and meso-level of networks, yielding a whole new set of traditional metrics – centrality indices, subgroup identification, density metrics etc – explicitly adapted to capture patterns in valued networks.

 

20 May, 2014 @ CEU

Szabolcs Számadó: The complexity of human cooperation and the five”rules

The complexity of human cooperation is unparalleled in nature. Humans readily cooperate with non-kin, form huge societies based on cooperation and their cooperative networks can reach across the globe. Cooperation can be classified into five categories: kin selection, direct reciprocity, indirect reciprocity, network reciprocity and group selection. I would like to discuss each type, indicate their strength and weaknesses as far as modelling concerned; and discuss their relevance to human cooperation. Last but not least I would like to highlight the link between gossip and reputation on the one hand and human cooperation on the other hand.

 

27 May @ MTA TK

Miklós Antal (UAB Barcelona): Limits to growth from a network perspective

Whether or not long-run economic growth can be sustainable has been debated for decades. Accumulated evidence supports two important conclusions. First, all major economies are dependent on growth: if GDP declines for a longer period, then significant social tensions can be expected due to rising unemployment, falling real wages and other problems. Second, further global economic growth is very likely to be accompanied by large-scale environmental degradation. Therefore, social and environmental sustainability are likely to be incompatible. To address this dilemma, strategies are needed to decouple – as much as possible – GDP reduction from social problems and GDP growth from environmental problems. In this exploratory talk, I will look at both correlations and the prospects of decoupling from a network perspective. This can be useful because both correlations are caused, at least in part, by network-related mechanisms. For instance, social relationships that force actors to adopt competitive strategies are an important reason for the correlation between GDP reduction and social tensions. Furthermore, the correlation between GDP and environmental impacts is persistent in several fields because policy effectiveness is limited by the increasing complexity of stakeholder networks. I will discuss network-related aspects of the growth debate and highlight the possible role of the quantification of network effects in devising sustainability strategies.

 

10 June @ MTA TK

Renáta Németh (ELTE): Discrete Graphical Models in Social Mobility Research

Graphical models provide a well visualizable and easily interpretable representation of complex associations. The models are based on graphs in which nodes represent variables and missing edges represent conditional independence statements. These models are well suited to model direct and indirect associations and effects that are of central importance in many problems of social mobility research.

The talk provides a unified view of many of the graphical models discussed in a largely scattered literature. The marginal loglinear modeling framework proposed relies on parameters that capture aspects of associations among the variables that are relevant for the graph and, depending on the substantive problem at hand, may lead to a deeper insight than other approaches.