Empirical modelling of contagion: a review of methodologies

There is now a reasonably large body of empirical work for the existence of contagion during financial. A range of different methodologies are in use, it difficult to assess the evidence for and against, and particularly its significance in transmitting between countries. The origins of current empirical of contagion stem from Sharpe (1964) and Grubel Fadner (1971), and more recently from King and (1990), Engle et al. (1990) and Bekaert and (1992). k aim of the present paper is to provide a unifying to highlight the key similarities and differences the various approaches. For an overview the literature see Dornbusch et al. (2000) and Pericoli Sbracia (2003). The proposed framework is based on latent factor structure which forms the basis of the of Dungey and Martin (2001), Corsetti et al. (2001, 2003) and Bekaert et al. (2005). This framework used to compare directly the correlation analysis popularized in this literature by Forbes and (2002), the VAR approach of Favero and (2002), the probability model of Eichengreen al. (1995, 1996) and the co-exceedance approach of et al. (2003).