Caracterização do mercado acionário brasileiro por meio dos cálculos de entropia, mutual information e a complexidade de Lempel-Ziv
We can describe financial markets as a Complex System, and the application of concepts from areas such as Econophysics and Information Theory provide a basis for understanding their characteristics and patterns. Thus, through the Entropy, Mutual Information (MI) and LempelZiv Complexity (CLZ) calcul...
| Autores: | , |
|---|---|
| Formato: | artículo |
| Estado: | Versión publicada |
| Fecha de publicación: | 2021 |
| País: | Brasil |
| Recursos: | Pontifícia Universidade Católica de São Paulo (PUC-SP) |
| Repositorio: | Redeca |
| Idioma: | portugués |
| OAI Identifier: | oai:ojs.pkp.sfu.ca:article/56637 |
| Acesso em linha: | https://revistas.pucsp.br/index.php/redeca/article/view/56637 |
| Access Level: | acceso abierto |
| Palavra-chave: | Sistemas Complexos Econofísica Mercados Financeiros Entropia Mutual Information Complexidade de Lempel-Ziv Complex Systems Ecophysics Financial Markets Entropy Lempel-Ziv Complexity |
| Resumo: | We can describe financial markets as a Complex System, and the application of concepts from areas such as Econophysics and Information Theory provide a basis for understanding their characteristics and patterns. Thus, through the Entropy, Mutual Information (MI) and LempelZiv Complexity (CLZ) calculations we sought to empirically characterize the series of daily stock returns on the São Paulo Stock Exchange between 2014 and 2018. From the results, it was seen that the most entropic days in the series were also those with variations in Ibovespa log returns and volume considerably higher than its daily average, reinforcing the fact that these are configured as days of less redundancy and greater informational content. For MI, the clusters were more delimited than those found via Pearson's Correlation. In CLZ, the low variability of substrings demonstrate a series with cyclic patterns and low complexity. |
|---|