|
on Network Economics |
| By: | Cabrales, Antonio; Hauk, Esther |
| Abstract: | In this paper we model the interaction between leaders, their followers and crowd followers in a coordination game with local interaction. The steady states of a dynamic best-response process can feature a coexistence of Pareto dominant and risk dominant actions in the population. The existence of leaders and their followers, plus the local interaction, which leads to clustering, is crucial for the survival of the Pareto dominant actions. The evolution of leader and crowd followership shows that leader followership can also be locally stable around Pareto dominant leaders. The paper answers the questions (i) which Leader should be removed and (ii) how to optimally place leaders in the network to enhance payoff dominant play. |
| Date: | 2022–07–12 |
| URL: | http://d.repec.org/n?u=RePEc:osf:socarx:8mect&r=net |
| By: | Francis X. Diebold (University of Pennsylvania); Kamil Yilmaz (Koc University) |
| Abstract: | We offer retrospective and prospective assessments of the Diebold-Yilmaz connectedness research program, combined with personal recollections of its development. Its centerpiece in many respects is Diebold and Yilmaz (2014), around which our discussion is organized. |
| Keywords: | Contagion, Spillovers, Financial markets, Vector autoregressions, Variance decompositions. |
| JEL: | C1 C3 G1 |
| Date: | 2022–11 |
| URL: | http://d.repec.org/n?u=RePEc:koc:wpaper:2207&r=net |
| By: | Dewenter, Ralf (Helmut Schmidt University, Hamburg); Löw, Franziska (Helmut Schmidt University, Hamburg) |
| Abstract: | In contrast to traditional business models, two-sided platforms internalize indirect network effects that exist between different groups of platform participants. The strength of the network effects has a decisive influence on the success of the platform and its market position. Markets with particularly strong network effects are also often characterized by a high degree of concentration. However, the strength of the network effects is not exogenously given but can be influenced by targeted investment. This paper analyses how platforms can affect network effects by investing in appropriate infrastructure, data, or artificial intelligence. We derive optimal quantities, prices, profits, and investments depending on different types of investments. |
| Keywords: | two-sided markets; indirect network effects; endogenous network effects; optimal investment strategy |
| JEL: | D21 D42 L10 |
| Date: | 2022–08–09 |
| URL: | http://d.repec.org/n?u=RePEc:ris:vhsuwp:2022_194&r=net |
| By: | Spiros Bougheas; Adam Hal Spencer |
| Abstract: | We introduce endogenous fire sales into a simple network model. For any given initial distribution of shocks across the network, we develop a clearing algorithm to solve for the financial equilibrium. We then utilise the results to perform ex ante risk assessment and derive risk premia for every balance sheet item where liabilities are differentiated according to priority rights. We find that risk premia reflect both idiosyncratic risk and risk of contagion (network risk). Moreover, we show that network risk magnifies the gap between the risk premia of equity and debt. We also perform comparative statics, showing that changes to the distribution of shocks and network structure can have substantial effects on the level of systemic losses. |
| Keywords: | Networks; Fire Sales; Systemic Risk Premia; Risk Assessment |
| Date: | 2022 |
| URL: | http://d.repec.org/n?u=RePEc:not:notcfc:2022/04&r=net |
| By: | Elena Farahbakhsh Touli; Hoang Nguyen; Olha Bodnar |
| Abstract: | In this paper, we study the connection between the companies in the Swedish capital market. We consider 28 companies included in the determination of the market index OMX30. The network structure of the market is constructed using different methods to determine the distance between the companies. We use hierarchical clustering methods to find the relation among the companies in each window. Next, we obtain one-dimensional time series of the distances between the clustering trees that reflect the changes in the relationship between the companies in the market over time. The method of statistical process control, namely the Shewhart control chart, is applied to those time series to detect abnormal changes in the financial market. |
| Date: | 2022–10 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2210.16679&r=net |
| By: | Zachary Fulker; Patrick Forber; Rory Smead; Christoph Riedl |
| Abstract: | Sending and receiving signals is ubiquitous in the living world. It includes everything from individual molecules triggering complex metabolic cascades, to animals using signals to alert their group to the presence of predators. When communication involves common interest, simple sender-receiver games show how reliable signaling can emerge and evolve to transmit information effectively. These games have been analyzed extensively, with some work investigating the role of static network structure on information transfer. However, no existing work has examined the coevolution of strategy and network structure in sender-receiver games. Here we show that coevolution is sufficient to generate the endogenous formation of distinct groups from an initially homogeneous population. It also allows for the emergence of novel ``hybrid'' signaling groups that have not previously been considered or demonstrated in theory or nature. Hybrid groups are composed of different complementary signaling behaviors that rely on evolved network structure to achieve effective communication. Without this structure, such groups would normally fail to effectively communicate. Our findings pertain to all common interest signaling games, are robust across many parameters, and mitigate known problems of inefficient communication. Our work generates new insights for the theory of adaptive behavior, signaling, and group formation in natural and social systems across a wide range of environments in which changing network structure is common. We discuss implications for research on metabolic networks, among neurons, proteins, and social organisms. |
| Date: | 2022–10 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2210.17309&r=net |
| By: | Aymanns, Christoph; Foerster, Jakob; Georg, Co-Pierre; Weber, Matthias (University of St. Gallen) |
| Abstract: | We propose multi-agent reinforcement learning as a new method for modeling fake news in social networks. This method allows us to model human behavior in social networks both in unaccustomed populations and in populations that have adapted to the presence of fake news. In particular the latter is challenging for existing methods. We find that a fake-news attack is more effective if it targets highly connected people and people with weaker private information. Attacks are more effective when the disinformation is spread across several agents than when the disinformation is concentrated with more intensity on fewer agents. Furthermore, fake news spread less well in balanced networks than in clustered networks. We test a part of these findings in a human-subject experiment. The experimental evidence provides support for the predictions from the model. This suggests that our model is suitable to analyze the spread of fake news in social networks. |
| Date: | 2022–07–21 |
| URL: | http://d.repec.org/n?u=RePEc:osf:socarx:y4mkd&r=net |
| By: | Pulgar, Javier; Ramírez, Diego; Candia, Cristian |
| Abstract: | People perceive, interpret, and assess others’ behaviors to build social ties even at young ages, affected by local contexts, which can lead, in the end, to improving learning experiences, resolving complex tasks, or creating companies. After the recent healthcare crisis due to COVID-19, online and hybrid social interactions are more prevalent in current social relationships and particularly relevant in education because information communication technologies now modulate behaviors. Here, we explore the role of social behaviors and their local contexts in building social networks and driving individual network centrality. Chilean education configuration provides a natural setting for exploring social relationships because students remain in the same group eight hours per day during their entire primary (eight years) and secondary (four years) school years. Thus, in the social mix of the classroom, students would show and witness their and others’ behaviors in both academic and social activities. We conduct two studies in different educational levels: i) primary school students (858 students from 45 different classrooms) and ii) secondary (191 students from eight different classrooms) to test cognitive and social drivers for network centrality. In the first study, we map the network collaboration for 3rd, 4th, and 5th-grade students. We find that 3rd-grade students exhibit more connections in their social networks than older students. Besides, we observe that social perception of popularity, prosociality, aggressiveness, and social preferences drive network centrality. In the second study, we separated secondary students into two different local contexts: hybrid and online teaching modalities. We find that students exhibit more network connectivity in the hybrid setting, and good student perception does not drive individual network centrality. On the other hand, in classrooms performing online teaching, the perception of good students is positively associated with network centrality. These results support the idea that students behave by following and interpreting social behaviors in their interactions, even in young populations. The availability and use of information for tie formation are dependable on students’ cognitive maturity and accessibility to face-to-face interactions, where popularity and pro-social behavior, for instance, become more evident. |
| Date: | 2022–06–15 |
| URL: | http://d.repec.org/n?u=RePEc:osf:socarx:yh8qc&r=net |
| By: | Hart, Jordan D. A.; Franks, Daniel Wayne; Brent, Lauren; Weiss, Michael N. |
| Abstract: | We present a user-friendly R package for fitting BISoN models and conducting a wide variety of social network analyses. The package uses a formula-based syntax that will be familiar to R users, and a single core function for fitting edge models that we hope will be accessible to practitioners. We hope our R package will facilitate wider adoption of the BISoN framework, enable more reliable inferences, and allow more sophisticated and specialised analysis of social networks. |
| Date: | 2022–11–08 |
| URL: | http://d.repec.org/n?u=RePEc:osf:osfxxx:ywu7j&r=net |
| By: | McCann, Mark (University of Glasgow); Mitchell, Kirstin; Broccatelli, Chiara; Purcell, Carrie; Simpson, Sharon; McDaid, Lisa; Elliott, Lawrie; Moore, Laurence |
| Abstract: | Background: The traditional focus of adolescent sexual behaviour research is on risk and adverse outcomes and on first vaginal intercourse as a portent of risk. This preoccupation with first sex limits research on social processes of sexual development and is out of step with the contemporary reality of young people’s early sexual experiences in which vaginal sex (if it happens at all) often follows a wide range of intimate and sexual experiences. Aim: With an innovative approach combining latent class analysis and social network analysis, we assessed how sexual knowledge, positive norm adherence, confidence and sexual activity are patterned across school-year social networks and explored the extent to which individualist and social influence theories could explain these patterns. Methods: We analysed cross-sectional survey data from 10 school year-group (students age 14-16) from the STis And Sexual Health (STASH) feasibility study. Students completed measures on sexual health knowledge, norms, confidence and behaviours, and named up to six friends in their year group. We used Latent Class Analysis to categorise patterns of sexual behaviour, and Exponential Random Graph Models to assess how sexual behaviour, knowledge, norms and confidence related to friendship ties. Results: Of 1,446 students 21% (n=309) did not report any sexual experience (inactive), 42% (602) reported some sexual experience but not oral or vaginal sex, and 22% (323) had oral and/or vaginal sex. Friendship ties were more likely between students who had similar levels of competencies (knowledge, norms and confidence). Friendships were also more likely between students who were sexually inactive (OR 1.73 95% CI 1.54, 1.95), and between students who reported oral/vaginal sex (OR 1.77 95% CI 1.46, 2.15). Active but not oral/vaginal sex students were slightly less likely (OR 0.86 95% CI 0.75, 0.99) to have friendship ties with each other. Interpretation: Our findings suggest that peer influence operating on sexual health competencies may occur independently of current sexual behaviour and there may not be strong naturally occurring social influence process to initiate pre-intercourse sexual behaviours. Our findings suggest that sexual health interventions acting on network influences are justified and that the focus of such interventions should shift from first intercourse to patterning of developing sexual repertoires. |
| Date: | 2022–10–15 |
| URL: | http://d.repec.org/n?u=RePEc:osf:osfxxx:upx27&r=net |
| By: | Sylvain Mignot; Annick Vignes (CAMS - Centre d'Analyse et de Mathématique sociales - EHESS - École des hautes études en sciences sociales - CNRS - Centre National de la Recherche Scientifique) |
| Abstract: | A fundamental assumption in economics is that rational individuals act in their own self interest. One implication is that, when trading, buyers are supposed to seek for the lowest price and sellers for the highest one and social interactions are not considered. It is now largely accepted that social relationships affect the efficiency of a market structure (centralized or decentralized) . The objectives of the current study is to examine the network structures of a very specific market : the Boulogne-sur-mer fish market. On this market two market design (auctions and bilateral exchanges) coexist, each beeing used by the same buyers and sellers, exchanging similar goods. For each sub-market we examine (1) the global network structure, (2) the local network structure, and (3) we identify the traders characteristics that best explain the network structures. The objective is to identify the the influence of trust nad the influence of reputation in the individual choices of trading partners. Structural measures are used to characterize networks structures. Exponential random graph models are used to evaluate how trader characteristics explain purchasing patterns, and how the influence of these characteristics vary with the market mechanism. We bring into the light that, when the transaction links on the auction market reflects the economic constraints of the partners, the relationships on the bilateral market depends on something more. Clearly, the prices of the bilateral transactions are the consequences of economics and non economics determinants. At first glance, the stable co-existence of two market structures looks like a paradox. Our results help to understand the distinctive characteristics and functioning of each sub-market. This discussion contributes to the debate about the efficiency of market structures. |
| Keywords: | social networks,market structures,trust,reputation,fish markets |
| Date: | 2022–11–08 |
| URL: | http://d.repec.org/n?u=RePEc:hal:journl:hal-03801324&r=net |
| By: | Schauf, Andrew; Oh, Poong |
| Abstract: | When populations share common-pool resources (CPRs), individuals decide how much effort to invest towards resource extraction and how to allocate this effort among available resources. We investigate these dual aspects of individual choice in networked games where resources undergo regime shifts between discrete quality states (viable or depleted) depending on collective extraction levels. We study the patterns of extraction that emerge on various network types when agents are free to vary extraction from each CPR separately to maximize their short-term payoffs. Using these results as a basis for comparison, we then investigate how results are altered if agents fix one aspect of adaptation (magnitude or allocation) while letting the other vary. We consider two constrained adaptation strategies: uniform adaptation, whereby agents adjust their extraction levels from all CPRs by the same amount, and reallocation, whereby agents selectively shift effort from lower- to higher-quality resources. A preference for uniform adaptation increases collective wealth on degree-heterogeneous agent-resource networks. Further, low-degree agents retain preferences for these constrained strategies under reinforcement learning. Empirical studies have indicated that some CPR appropriators ignore—while others emphasize—allocation aspects of adaptation; our results demonstrate that structural patterns of resource access can determine which behavior is more advantageous. |
| Date: | 2021–11–08 |
| URL: | http://d.repec.org/n?u=RePEc:osf:socarx:wmtqk&r=net |