Contagion among major world markets: a wavelet approach

Abstract

PurposeThe purpose of this paper is to examine contagion among the major world markets during the last 25 years and propose a new way to analyze contagion with wavelet methods.Design/methodology/approachThe analysis uses a novel way to study contagion using wavelet methods. The comparison is made between co‐movements at different time scales. Co‐movement methods of the discrete wavelet transform and the continuous wavelet transform are applied.FindingsClear signs of contagion among the major markets are found. Short time scale co‐movements increase during the major crisis while long time scale co‐movements remain approximately at the same level. In addition, gradually increasing interdependence between markets is found.Research limitations/implicationsBecause of the chosen method, the approach is limited to large data sets.Practical implicationsThe research has practical implications to portfolio managers etc. who wish to have better view of the dynamics of the international equity markets.Originality/valueThe research uses novel wavelet methods to analyze world equity markets. These methods allow the markets to be analyzed in the whole state space.

Shari’ah screening, market risk and contagion: A multi-country analysis

“…This common cause is independent from shocks, with the exclusion of the case of the financial crisis. By using a wavelet method, Ranta (2013), we found that several times in the last twenty-five years, contagion was a major factor in the major world markets and strong signs of contagion could be seen during the global financial crisis of 2008-2009. (Rua and Nunes, 2009;Ranta, 2013).…”

“…By using a wavelet method, Ranta (2013), we found that several times in the last twenty-five years, contagion was a major factor in the major world markets and strong signs of contagion could be seen during the global financial crisis of 2008-2009. (Rua and Nunes, 2009;Ranta, 2013). Moreover, the wavelet coherence used in our study fits the analysis of co-movement between economic variables.…”

Time–Scale Relationship between Securitized Real Estate and Local Stock Markets: Some Wavelet Evidence

“…Despite much academic effort regarding the time-frequency relationships across stock markets using continuous wavelet transform (CWT) (e.g., Aguiar-Conraria et al 2008;Loh 2013;Dima et al 2015;and Tiwari et al 2016) and maximal overlap discrete wavelet transform (MODWT) (e.g., Ranta 2013;Deora and Nguyen 2013;and Dewandaru et al 2016) since the last decade, previous time-scale studies on real estate markets, as well as between real estate and stock markets, are lacking, with the exceptions of Zhou (2010Zhou ( , 2012, Li et al (2015), and Liow et al (2018aLiow et al ( , 2018b. This knowledge gap will be filled by the present study.…”

“…On the second issue relating to financial contagion, since any increase in the real estate and stock co-movement from the pre-crisis period to the crisis period may take the form of interdependence or contagion, there is the suspicion that the observed co-movement between securitized real estate and local stock markets is caused by contagion in the context of globalization. Whilst some researchers such as Rua and Nunes (2009) and Ranta (2013) presume that contagion is a temporary increase of short time-scale co-movement (or "pure" contagion, Dornbusch et al (2000)), we also consider whether the detected real estate-stock co-movement is also affected by "fundamentals-based" contagion (Kaminsky and Reinhart 2000). "Pure" contagion describes a situation where an excessive transmission of shocks due to a crisis is beyond any idiosyncratic disturbances and fundamental linkages, and is relatively fast and disappears in a short period.…”

“…However, the chosen filter (the Haar filter) and the window sizes do not appear to have any significant effect on the results. 7 Ranta (2013) used the same methodology in investigating contagion across major stock markets. To confirm the presence of contagion (or otherwise), a paired-sample t-test evaluates if there is a significant increase in the wavelet correlation coefficient after the two crises.…”