A New Class of Tests of Contagion With Applications

Fry, Renée; Martin, Vance L.; Tang, Edward Chi Ho

doi:10.1198/jbes.2010.06060

Cited by 163 publications

(183 citation statements)

References 28 publications

Supporting

Mentioning

173

Contrasting

Unclassified

Order By: Relevance

“…Forbes and Rigobon (2002) show that the presence of heteroscedasticity biases this type of testing procedure, leading to over-acceptance of the hypothesis of the presence of contagion. Bae et al (2003), Pesaran and Pick (2007) and Fry et al (2010) propose testing procedures robust to the presence of heteroscedasticity.…”

Section: Introductionmentioning

confidence: 99%

Modelling Financial Markets Comovements during Crises: A Dynamic Multi-Factor Approach

Belvisi¹,

Pianeti²,

Urga

2016

Dynamic Factor Models

View full text Add to dashboard Cite

paricular J. Breitung, M. Hallin and M. Marcellino, for useful discussions and valuable comments. Special thanks to the Editors, Eric Hillebrand and Siem Jan Koopman, and two anonymous referee for very helpful comments and suggestions that greatly helped to improve the paper. Riccardo Borghi has provided very insightful comments on a previous version of the paper. The usual disclaimer applies. Riccardo Pianeti acknowledges …nancial support from the Centre for Econometric Analisis at Cass and the EAMOR Doctoral Programme at Bergamo University.y KNG Securities, London (UK). We propose a novel dynamic factor model to characterise comovements between returns on securities from di¤erent asset classes from di¤erent countries. We apply a global-classcountry latent factor model and allow time-varying loadings. We are able to separate contagion (asset exposure driven) and excess interdependence (factor volatility driven). Using data from 1999 to 2012, we …nd evidence of contagion from the US stock market during the 2007-09 …nancial crisis, and of excess interdependence during the European debt crisis fromMay-2010 onwards. Neither contagion nor excess interdependence is found when the average measure of model implied comovements is used.JEL: C3, C5, G1.

show abstract

Section: Introductionmentioning

confidence: 99%

Modelling Financial Markets Comovements during Crises: A Dynamic Multi-Factor Approach

Belvisi¹,

Pianeti²,

Urga

2016

Dynamic Factor Models

View full text Add to dashboard Cite

show abstract

“…We find strong evidence of a negative association between the level of our sentiment index and the risk of contagion for the 24 In order to avoid double counting the sentiment factor, the residuals from an asset pricing model without the investor sentiment variable are used in this analysis. 25 As suggested by 'noise trader' theories, sentiment should bias the behavior of irrational investors but it is not ruled out that institutional investors would also be affected.…”

Section: Behavioral Mechanismsmentioning

confidence: 99%

Contagion Channels between Real Estate and Financial Markets

Hoesli

Reka

2013

SSRN Journal

View full text Add to dashboard Cite

The recent crisis has demonstrated the close linkages between various asset classes within a country as well as the association between assets internationally. The aim of this research is to provide for a better understanding of some of these linkages by conducting an empirical investigation of the channels underlying the risk of contagion between real estate and financial markets in the United States. We test for three financial mechanisms driving contagion: Information correlation, liquidity correlation, and portfolio rebalancing. A behavioral dimension in the crisis propagation is also examined by considering investor sentiment and panic risk in our analyses. A methodology based on quantile regressions and copulas is utilized for carrying out those tests. We find that contagion prevails between real estate and financial markets and that this is driven by behavioral and liquidity mechanisms. The correlated information and portfolio rebalancing hypotheses are strongly rejected. The results of this research should prove useful to investors seeking to hedge against market crashes as well as to policy makers who aim to mitigate such stressful episodes.JEL Classification: G01; G12; G02; G14 Executive SummaryThe connections between various assets within a country as well as the linkages between international markets have received considerable attention in the academic literature. This topic is of importance since it has direct implications for portfolio diversification and countries' financial stability. Financial contagion is a crucial issue related to this topic and its consequences on a country's economy may be tremendous, as illustrated by the recent financial crisis. Indeed, several markets suffered catastrophic losses initially triggered by large defaults by subprime borrowers in the U.S. mortgage markets. Loosely speaking, contagion can be defined as a rapid shock spillover that increases cross-market linkages. Examining the determinants of those linkages is crucial in order to fully understand this phenomenon. Despite the role played by real estate in the recent financial crisis and its importance in the economy, there is no study in the literature that addresses the issues of the contagion channels involving this market. This paper aims to fill this gap. More precisely, we use the interdependences between the U.S. real estate and equity markets as our testingground. Thus, we analyze the dynamics of financial contagion within a cross-asset framework (domestic contagion). Real estate investments trust data have been selected to proxy for real estate returns. As for financial markets, three sub-indices of equity markets (i.e. small cap, large cap and commercial bank stocks) are used for the analyses.In this study, we seek to answer two questions. First, defining contagion as return comovements that cannot be explained by economic fundamentals, is there contagion between real estate and financial markets? Second, what are the mechanisms underlying contagion in this context? We test for three financial ...

show abstract

“…Transmissions of these severe financial shocks from one assets market to another are always far exceeding expectations compared with the market linkage and the dependence structure in the normal period [4]. Forms of the unexpected change include the change in the codependence structures across financial markets, unusual correlation and also another additional crisis transmitting channels such as higher order co-moments of the assets returns [5]. The difference in volatility has been researched in [1].…”

Section: Introductionmentioning

confidence: 99%

“…[2] Focused on the non-linear relationship. A class of new testing method was developed by [5], [6], [7] and [8] which used higher order comoments to test the properties of the distributions of assets return.…”

Section: Introductionmentioning

confidence: 99%

“…The definition of contagion phenomenon: a significant change in cross-market joint comoment after a financial shock [8] has been applied. The paper measures whether there was such a financial contagion from the US to Australian financial firms during the financial crisis during the period of 2007-2014 which consists of the US subprime mortgage crisis, global financial crisis (also called the Great Recession) and the European debt crisis by using the method of joint test, which consists of linear and higher order co-moments such as: cross-market correlation [3], cross-market skewness (or co-skewness: the relationship between the asset return in market i and the return volatility in market j) [5], cross-market kurtosis (or co-kurtosis: the relationship between the asset return in market i and return skewness in market j) and cross-market volatility (or co-volatility: the relationship between the return volatility of markets i and j) [7]. The higher order co-moments are derived by assuming the distribution of markets stock return as a bivariate normal distribution with the higher order moments and co-moments, and the normality is certified by calculating the Lagrange Multiplier statistic.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation