The Impact of Network Connectivity on Factor Exposures, Asset Pricing and Portfolio Diversification

Billio, Monica; Caporin, Massimiliano; Panzica, Roberto; Pelizzon, Loriana

doi:10.2139/ssrn.2914218

Cited by 17 publications

(19 citation statements)

References 63 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In this enterprise, we take inspiration from a recent strand of spatial econometrics literature that includes finance-based weight matrices in their analysis (Keiler and Eder (2013), Blasques et al (2016), Tonzer (2015), and Billio et al (2017)), and starts to introduce the network terminology in the discussion. However, building on Caporin and Paruolo (2015) differentiates us from the latter papers.…”

Section: A Model For Spatial Econometrics Of Networked Riskmentioning

confidence: 99%

“…The second ingredient is our consideration of the weight matrix as a directed network. Directedness is not present in classical approaches, but is already considered in Billio et al (2017);Tonzer (2015) and Blasques et al (2016).…”

Section: A Model For Spatial Econometrics Of Networked Riskmentioning

confidence: 99%

“…The intersection and interaction of different research fields, such as network science, spatial econometrics, economics, and finance, give rise to new developments and tools. In particular, one of the most advanced strands of the spatial econometrics literature (see Keiler and Eder (2013); Blasques et al (2016); Caporin and Paruolo (2015); Tonzer (2015); Billio et al (2017)) focuses on statistical, economic, and financial relationships. Our work belongs to this strand of the literature and aims at building a bridge between contagion, volatility spillover, and networks.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Networks in Risk Spillovers: A Multivariate GARCH Perspective

et al. 2018

Self Cite

View full text Add to dashboard Cite

We propose a spatiotemporal approach for modeling risk spillovers using time-varying proximity matrices based on observable financial networks and introduce a new bilateral specification. We study covariance stationarity and identification of the model, and analyze consistency and asymptotic normality of the quasi-maximum-likelihood estimator. We show how to isolate risk channels and we discuss how to compute target exposure able to reduce system variance. An empirical analysis on Euro-area cross-country holdings shows that Italy and Ireland are key players in spreading risk, France and Portugal are the major risk receivers, and we uncover Spain's non-trivial role as risk middleman.

show abstract

Section: A Model For Spatial Econometrics Of Networked Riskmentioning

confidence: 99%

Section: A Model For Spatial Econometrics Of Networked Riskmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Networks in Risk Spillovers: A Multivariate GARCH Perspective

et al. 2018

Self Cite

View full text Add to dashboard Cite

show abstract

“…To shed further light on the differences between the composite networks and generate some evidence of the improvement gained by the linear factor model augmented with network combination, we provide descriptive analyses -in Tables 6 to 8 achieved by the latter, over the multifactor model, was already documented in (Billio et al, 2015). Table 6: The table reports residual correlation descriptive analyses for the Banks dataset.…”

Section: The Composite Causality Networkmentioning

confidence: 99%

Estimation and Model-Based Combination of Causality Networks

2017

Self Cite

View full text Add to dashboard Cite

Statistical tests based on Granger causality are used in finance to detect the links among financial institutions that might be interpreted as channels along which shocks spread through the financial system. The links' structure can be formally defined as a network. Despite the growing interest in the financial network topic and the increasing understanding that there might be different channels over which financial contagion spreads, the literature on combinations of financial or economic networks is still very limited. In fact, the available competing approaches to estimate networks among financial institutions suggest the coexistence of different channels for the spread of risk. It is therefore of fundamental importance to allow for the possibility of combining those alternative risk spreading channels to obtain a more complete picture of risk propagation within the financial system. Furthermore, when focusing on the approaches for estimating financial company links, we believe that Granger causality should be contrasted with methods pointing in a more clear way to the risk dimension, for instance by detecting causality among quantiles. Therefore, we utilize parametric and notparametric approaches for the estimation on the networks based on quantile causality tests.We show how to use a linear factor model as a device for estimating a combination of several networks that monitor the links across variables from different viewpoints, and we demonstrate that Granger causality test should be combined with quantile-based causality when the focus is on risk propagation. We empirically validate our two main proposals concerning the use of quantile causality to infer the network structure across a set of (financial) variables, and the model-based combination of causality networks. By using different datasets (US industrial portfolio returns, and a set of large banks and insurance companies), we first provide evidence of the different network structures that we can estimate from Granger causality and quantile causality. We then show how the networks differ across methods and over two different samples relating to the global financial crisis (2006)(2007)(2008) and to the years 2011-2015. Our results suggest that quantile causality networks are denser than Granger causality networks, a finding of relevance to systemic risk interpretation, because a denser network is indicative of a much larger set of links, and thus explains a possibly greater systemic effect of shocks.Electronic copy available at: https://ssrn.com/abstract=2909585 Abstract Causality is a widely-used concept in theoretical and empirical economics. The recent financial economics literature has used Granger causality to detect the presence of contemporaneous links between financial institutions and, in turn, to obtain a network structure. Subsequent studies combined the estimated networks with traditional pricing or risk measurement models to improve their fit to empirical data. In this paper, we provide two contributions: we show how to use a linear facto...

show abstract

“…Pour une revue complète de la littérature sur les mesures de risque systémique, voir par exemple Bisias et al [2012] ; De Bandt et al [2015] et Benoit et al [2017a]. Voir aussi, plus spécifiquement, Abergel et al [2013] ; Diebold et Yilmaz [2014] ; Hautsch et al [2015] et Billio et al [2017] pour la littérature sur les réseaux servant à quantifier le risque systémique. temps réel.…”

Section: Introductionunclassified

Du MEDAF avec risque systémique à la détermination des institutions financières d’importance systémique

Garibal¹,

Kouontchou

Maillet

2018

Revue Économique

View full text Add to dashboard Cite

Nous nous proposons de tester dans cet article une extension du MEDAF, dans lequel nous ajoutons un facteur de risque systémique, mesuré par une agrégation des principales mesures de risques systémiques, comme proposée récemment par différents auteurs, dans le cadre d'une Analyse en Composantes Principales Parcimonieuse (ACPP). Il ressort de nos analyses empiriques menées sur le marché américain que le risque systémique est effectivement une composante importante de la rémunération sur certains titres. Nous proposons finalement une application originale servant à l'identification et au classement des Institutions Financières d'Importance Systémique (IFIS). From the CAPM with Systemic risk to the designation of the Systemically Important Financial Institutions We propose herein to test an extension of the traditional CAPM, in which is added a factor of systemic risk, build thanks to a Sparse Principal Component Analysis (SPCA) of a large set of systemic risk measures, as recently proposed by several authors. Empirical tests show that, on the American equities market, the new systemic risk factor is highly significant when pricing assets. At the end, we propose an original application in order to detect and rank Systemically Important Financial Institutions (SIFI).

show abstract

The Impact of Network Connectivity on Factor Exposures, Asset Pricing and Portfolio Diversification

Cited by 17 publications

References 63 publications

Networks in Risk Spillovers: A Multivariate GARCH Perspective

Networks in Risk Spillovers: A Multivariate GARCH Perspective

Estimation and Model-Based Combination of Causality Networks

Du MEDAF avec risque systémique à la détermination des institutions financières d’importance systémique

Contact Info

Product

Resources

About