Aplicabilidade de técnicas de redução de dimensionalidade no índice Bovespa (Ibovespa B3).

Abstract

Dimensionality is known as the number of input variables or features present in a data space, and the process of addressing the number of features representing that space is called dimensionality reduction. Dimensionality reduction aims to convert a data space of a certain dimension into one that is lower in order to ensure that it still provides equivalent information. These techniques are widely used in machine learning modeling to obtain a suitable predictive model while solving classification and regression problems. In this work, the implementation of some dimensionality reduction techniques, on the main stock index of the Brazilian stock exchange, the Bovespa index (Ibovespa B3), was carried out with the aid of the Python programming language.