Diego Nascimento, ICMC/USP
This paper aimed to promote the use of the Dynamic Conditional Correlation GARCH (DCC-GARCH) mutation model, by using a Monte Carlo approach via Markov chains in the estimation of parameters, as well as visually demonstrate this time-dependence variation. The empirical reflection fell on some aspects that explain the relationship between the indexes of the world’s markets. Fifteen indexes were analyzed from the main financial markets, from developed and developing countries from different continents. The first technique implemented was Visual Data Mining, which aided in clarifying the research problem and was able to identify simultaneous temporal patterns. Then, Dynamic Conditional Correlation of the markets’ returns, known as DCC-GARCH, under the Bayesian Approach is introduced. The study showed a presence of financial contagion suggesting structural changes in the joint evolutions of the world financial market. The findings revealed in this research suggest that the performance of indices are similar, with a joint evolution, as an example the mutual increase from 2010 and 2012. Finally, most index returns, especially SPX and NDX, evolve over time with a higher positive correlation, that is, they contains the same type of development. Authors: Diego Nascimento, Cleber Xavier, Israel Felipe, Francisco Louzada.