Vector autoregressive models have widely been applied in macroeconomics and macroeconometrics to estimate economic relationships and to empirically assess theoretical hypothesis. To achieve the latter, we propose a Bayesian inference approach to analyze the dynamic interactions among macroeconomics variables in a graphical vector autoregressive model. The method decomposes the structural model into multivariate autoregressive and contemporaneous networks that can be represented in the form of a directed acyclic graph. We then simulated the networks with an independent sampling scheme based on a single-move Markov Chain Monte Carlo (MCMC) approach. We evaluated the efficiency of our inference procedure with a synthetic data and an empirical assessment of the business cycles hypothesis.
Peer-to-Peer (P2P) FinTech platforms allow cost reduction and service improvement in credit lending. However, these improvements may come at the price of a worse credit risk measurement, and this can hamper lenders and endanger the stability of a financial system. We approach the problem of credit risk for Peer-to-Peer (P2P) systems by presenting a latent factor-based classification technique to divide the population into major network communities in order to estimate a more efficient logistic model. Given a number of attributes that capture firm performances in a financial system, we adopt a latent position model which allow us to distinguish between communities of connected and not-connected firms based on the spatial position of the latent factors. We show through empirical illustration that incorporating the latent factor-based classification of firms is particularly suitable as it improves the predictive performance of P2P scoring models.
Vector autoregressive models have widely been applied in macroeconomics and macroeconometrics to estimate economic relationships and to empirically assess theoretical hypothesis. To achieve the latter, we propose a Bayesian inference approach to analyze the dynamic interactions among macroeconomics variables in a graphical vector autoregressive model. The method decomposes the structural model into multivariate autoregressive and contemporaneous networks that can be represented in the form of a directed acyclic graph. We then simulated the networks with an independent sampling scheme based on a single-move Markov Chain Monte Carlo (MCMC) approach. We evaluated the efficiency of our inference procedure with a synthetic data and an empirical assessment of the business cycles hypothesis.
Following the 2007-08 Global Financial Crisis, there has been a growing research interest on the spatial interrelationships between house prices in many countries. This paper examines the spatio-temporal relationship between house prices in the twelve provinces of the Netherlands using a recently proposed econometric modelling technique called the Bayesian Graphical Vector Autoregression (BG-VAR). This network approach is suitable for analysing the complex spatial interactions between house prices. It enables a data-driven identification of the most dominant provinces where temporal house price shocks may largely diffuse through the housing market. Using temporal house price volatilities for owner-occupied dwellings from 1995Q1 to 2016Q1, the results show evidence of temporal dependence and house price diffusion patterns in distinct sub-periods from different provincial housing sub-markets in the Netherlands. In particular, the results indicate that Noord-Holland was most predominant from 1995Q1 to 2005Q2, while Drenthe became most central in the period 2005Q3-2016Q1.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.