Análise da previsão de retorno dos principais fundos de renda fixa brasileira, através de algorítmos de aprendizagem de máquina
The financial market has an increasing number of investment funds, becoming more accessible to the general public, with its infinite possibilities and low entry values. At the same time with so many open opportunities, it has the individual peculiarities of each one of them, complexity in names, pro...
| Autor principal: | Raphael, Thiago Proença Meirelles Correa |
|---|---|
| Formato: | Trabalho de Conclusão de Curso (Graduação) |
| Idioma: | Português |
| Publicado em: |
Universidade Tecnológica Federal do Paraná
2021
|
| Assuntos: | |
| Acesso em linha: |
http://repositorio.utfpr.edu.br/jspui/handle/1/26140 |
| Tags: |
Adicionar Tag
Sem tags, seja o primeiro a adicionar uma tag!
|
| Resumo: |
The financial market has an increasing number of investment funds, becoming more accessible to the general public, with its infinite possibilities and low entry values. At the same time with so many open opportunities, it has the individual peculiarities of each one of them, complexity in names, profitability, variations and rates. Making it very difficult for the population to understand the same and understand how movements can occur based on the current economy. In this context, the present work aimed to analyze two investment funds and verify the return forecast through machine learning algorithms, related to three major national indicators: Exchange, CDI and IPCA. For this purpose, the Naive Bayes algorithm was used in the Weka tool (Waikato Environment for Knowledge Analysis), Linear Regression via Excel and the Forecast Worksheet via Excel. Based on the results for the years 2018, 2019 and 2020 . The results obtained were that there is no relationship between the variation of funds and the determined predictors, showing that the forecast by time series is still the most effective, even if far from a good result. |
|---|