Deep learning para el pronóstico de la dirección del USDCOP
Currently, forex is one of the assets with highest traded volume in Colombia. Conventional methodologies are commonly used to forecast exchange rates. However, a new research trend using artificial intelligence has emerged. There is a lack of research in USDCOP exchange rate forecasting using deep l...
- Autores:
-
Gutiérrez Salamanca, Paula Marcela
- Tipo de recurso:
- Fecha de publicación:
- 2020
- Institución:
- Universidad de los Andes
- Repositorio:
- Séneca: repositorio Uniandes
- Idioma:
- spa
- OAI Identifier:
- oai:repositorio.uniandes.edu.co:1992/51009
- Acceso en línea:
- http://hdl.handle.net/1992/51009
- Palabra clave:
- Cambio exterior
Aprendizaje automático (Inteligencia artificial)
Análisis de series de tiempo
Ingeniería
- Rights
- openAccess
- License
- https://repositorio.uniandes.edu.co/static/pdf/aceptacion_uso_es.pdf
| Summary: | Currently, forex is one of the assets with highest traded volume in Colombia. Conventional methodologies are commonly used to forecast exchange rates. However, a new research trend using artificial intelligence has emerged. There is a lack of research in USDCOP exchange rate forecasting using deep learning in Colombia. Therefore, the main goal of this study is to predict the direction of change for the USDCOP with deep learning methodologies so it can be used in trading models to obtain higher returns on investment portfolios. Results demonstrate that deep neural networks outperformed the forecasting accuracy of conventional models such as ARIMA in different periods of time and time-steps.--Taken from the Degree Document Format. |
|---|