Dynamic correlation analysis in the ASEAN equity markets during 2009–2018

  • 110 Views
  • 13 Downloads

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License

This study examines the static and dynamic correlations in the ASEAN equity markets. The importance of this research appears from the fact that practitioners can get the benefit if their investments yield the same or higher returns given lower or the same risk in their portfolio. Firstly, this advantage comes from including the assets that decrease volatility of the portfolio. Hence, the correlation between the ASEAN markets should be examined. Secondly, co-movements in market realizations may increase global financial instability. Its existence is important for international investors, financial institutions, and policy makers. The study locates the relationship between ASEAN and its major trading partners, including Japanese, US, and UK markets, in order to find more rational results. This study utilizes alternative multivariate GARCH forms to provide useful information on the dynamic evolution and implications of return volatilities. The results show that the volatilities of all the equity markets under study are persistent over time. The estimates from VEC model indicate that the movements of the US and UK equity market returns have some degree of influence on several of the ASEAN equity markets. The results imply that, first, most of the developing ASEAN equity markets work by its own information with small relation to the developed world. Second, it is still convincing to state that investing in ASEAN equity markets should provide investors a better mean-variance portfolio. And, third, buy-and-hold strategy seems to be more beneficial than readjusting the ASEAN equities portfolio.

view full abstract hide full abstract
    • Table 1. Stock market index descriptive statistics
    • Table 2. Unconditional correlation matrix of market return
    • Table 3. Johansen’s test statistics for cointegration rank
    • Table 4. Vector error corrections model estimates and analysis
    • Table 5. Univariate generalized autoregressive conditional heteroskedasticity (GARCH (1,1)) estimates
    • Table 6. Conditional correlation generalized autoregressive conditional heteroskedasticity (CCC-GARCH(1,1)) estimates
    • Table 7. Dynamic correlation generalized autoregressive conditional heteroskedasticity (DCC-GARCH(1,1)) estimates (mean of correlations)
    • Table 8. Dynamic correlation generalized autoregressive conditional heteroskedasticity (DCC-GARCH(1,1)) estimates (maximum of correlations)
    • Table 9. Dynamic correlation generalized autoregressive conditional heteroskedasticity (DCC-GARCH(1,1)) estimates (minimum of correlations)
    • Table 10. Dynamic correlation generalized autoregressive conditional heteroskedasticity (DCC-GARCH(1,1)) estimates (standard deviation of correlations)