|
Activity Number:
|
478
- Scalable Bayesian Models for Time Series and Dynamic Networks: Making an Impact in Business and Socio-Economic Applications
|
|
Type:
|
Invited
|
|
Date/Time:
|
Wednesday, July 31, 2019 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
|
Abstract #300157
|
Presentation
|
|
Title:
|
A Bayesian Approach to Trajectory-Based Longitudinal Networks, with Application to the European Interbank Market
|
|
Author(s):
|
Antonietta Mira* and Federica Bianchi and Stefano Peluso and Francesco Bartolucci
|
|
Companies:
|
Università della Svizzera italiana and Università dell'Insubria and Univesità della Svizzera italiana and Cattolica University and Università della Svizzera italiana and University of Perugia
|
|
Keywords:
|
Financial networks;
Dyadic exchange;
Latent trajectories;
MCMC algorithms;
Reciprocal exchange;
Global Financial Crises
|
|
Abstract:
|
Financial markets bring buyers and sellers together on the trading floor by revealing distinct patterns of exchange over time. To study the temporal evolution of such dyadic exchange patterns from a network perspective, we propose a Bayesian longitudinal extension of the modeling framework for relational data by Holland and Leinhardt (1981). The model focuses on dyadic patterns of exchange and assumes a log-linear specification based on individual main effects and time-specific second-order effects. As output from the model, we obtain individual trajectories for the tendency to relate with any other node, and overall trajectories for the tendencies to connect with others and reciprocate their actions. We estimate the model by an appropriate MCMC algorithm. The proposed approach is illustrated through the analysis of the temporal evolution of individual trading behaviors in the European interbank market over a period of ten years including the Global Financial Crisis.
|
Authors who are presenting talks have a * after their name.
