The Study of Nonlinear Correlation between Shanghai, Hongkong and American Stock Returns An Empirical Analysis Based on MS-VAR Model and MS-DCC-MVGARCH Model

Zhou, Decai; Lu, Xin; Liu, Lei; Fu, Bixuan

doi:10.2139/ssrn.2374713

Cited by 1 publication

(4 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As stated, the methodology used in this paper to determine regime changes within our data follows the work of Zhou et al (2014), who combine MS mechanisms with VAR models to examine correlations between Hong Kong, Shanghai and US stock markets, finding regimes for bull and bear markets between 2005 and 2013.…”

Section: Methodsmentioning

confidence: 99%

“…This MSVAR approach has previously been implemented successfully by Brandt and Freeman (2012) who use the technique to determine crisis periods within datasets relating to the violent conflict between Israel and Palestine. In the literature on financial contagion, Zhou et al (2014) construct MSVAR models to measure the nonlinear correlation between stock returns in Shanghai, Hong Kong and America, finding differing characteristics in the correlations amongst markets and various dynamic causal relationships. The MSVAR approach adopted in this paper then, also takes inspiration from the work of Zhou et al (2014).…”

Section: Analysis Of the Hong Kong And Tokyo Stock Marketsmentioning

confidence: 99%

“…In the literature on financial contagion, Zhou et al (2014) construct MSVAR models to measure the nonlinear correlation between stock returns in Shanghai, Hong Kong and America, finding differing characteristics in the correlations amongst markets and various dynamic causal relationships. The MSVAR approach adopted in this paper then, also takes inspiration from the work of Zhou et al (2014).…”

Section: Analysis Of the Hong Kong And Tokyo Stock Marketsmentioning

confidence: 99%

“…As mentioned, this paper will employ an Markov-Switching Bayesian Vector AutoRegression (MSBVAR) methodology largely developed by Brandt and Freeman (2012) and in line with the approach taken by Rodriguez (2007) and Zhou et al (2014). This paper adds to the existing literature on financial contagion by examining market correlations over varying frequencies of stock market data.…”

Section: Analysis Of the Hong Kong And Tokyo Stock Marketsmentioning

confidence: 99%

See 3 more Smart Citations

Crisis determination and financial contagion: an analysis of the Hong Kong and Tokyo stock markets using an MSBVAR approach

Troug¹,

Murray

2020

JES

View full text Add to dashboard Cite

PurposeThe purpose of this paper then, is to add to the existing literature on financial contagion. While a vast amount of the debate has been made using data from the late 1990s, this paper differentiates itself by analysing more current data, centred around the most recent global financial crisis, with specific focus on the stock markets of Hong Kong and Tokyo.Design/methodology/approachEmploying Pearson and Spearman correlation measures, the dynamic relationship of the two markets is determined over tranquil and crisis periods, as specified by an Markov-Switching Bayesian Vector AutoRegression (MSBVAR) model.FindingsThe authors find evidence in support of the existence of financial contagion (defined as an increase in correlation during a crisis period) for all frequencies of data analysed. This contagion is greatest when examining lower-frequency data. Additionally, there is also weaker evidence in some data sub-samples to support “herding” behaviour, whereby higher market correlations persist, following a crisis period.Research limitations/implicationsThe intention of this paper was not to analyse the cause or transmission mechanism of contagion between financial markets. Therefore future studies could extend the methodology used in this paper by including exogenous macroeconomic factors in the MSBVAR model.Originality/valueThe results of this paper serve to explain why the debate of the persistence and in fact existence of financial contagion remains alive. The authors have shown that the frequency of a time series dataset has a significant impact on the level of observed correlation and thus observation of financial contagion.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Analysis Of the Hong Kong And Tokyo Stock Marketsmentioning

confidence: 99%