A carregar...
Publicação
Data Science for finance: automated investment recommendation with python
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 1.31 MB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
Most investors have difficulty finding good companies to invest in the stock market. As there are several stock options, it is hard to monitor the companies’ performance. So, this thesis aims to backtest to validate some investments strategies. The backtest includes fixed income and variable income investments, but specifically stock investments.
The analyses are made in the Brazilian market. For fixed income, we calculate the profitability obtained from investments in treasury bonds, using the inflation and the basic rate of the economy as indexers.
For variable income, we test some strategies of stock selection. We analyse Joel Greenblatt magic formula and Ben Graham’s formula for choosing stocks. We also try to create a model to select stocks based on the quote and fundamental analyses.
This project aims to automate the selection of stocks based on historical data and fundamental indicators. It does not claim to be a generic model, as it would be unfeasible since there are several points of view according to different investor’s profiles.
The project also shows where reads can get data to create a model. Then the reader can use their market knowledge to modify the model and thus create a model that suits the reader's preference. This research will be based on the Brazilian market but may be expanded by the reader of other markets.
This study is for long-term investors and not for day traders that need different tools and analysis.
Descrição
Dissertation presented as the partial requirement for obtaining a Master's degree in Statistics and Information Management, specialization in Risk Analysis and Management
Palavras-chave
Financial market prediction Investment Stock Market Times Series Data Science