A deep learning approach to forecasting EUA future contracts
This directed research project explores the use of deep learning techniques to forecast the price of European Union Allowance (EUA) future contracts. In order to achieve this, a comprehensive review of the literature surrounding emission markets, trading schemes and long short-term memory neural net...
- Autores:
-
Villamizar Díaz, Jorge Hernando
- Tipo de recurso:
- Fecha de publicación:
- 2020
- Institución:
- Universidad de los Andes
- Repositorio:
- Séneca: repositorio Uniandes
- Idioma:
- eng
- OAI Identifier:
- oai:repositorio.uniandes.edu.co:1992/51454
- Acceso en línea:
- http://hdl.handle.net/1992/51454
- Palabra clave:
- Comercio de derechos de emisión
Mercado de futuros
Aprendizaje automático (Inteligencia artificial)
Administración
- Rights
- openAccess
- License
- http://creativecommons.org/licenses/by-nc-nd/4.0/
| Summary: | This directed research project explores the use of deep learning techniques to forecast the price of European Union Allowance (EUA) future contracts. In order to achieve this, a comprehensive review of the literature surrounding emission markets, trading schemes and long short-term memory neural networks is made. Possible explanatory variables for the EUA futures price are selected, curated, and the methodology used to get a forecast is described. Additionally, under the Anatolyev-Gerko (EP) test statistic and the Pesaran-Timmermann (DA) test, there is evidence that a long-short trading strategy using the forecasted values beats the random walk, in other words, the strategy generates value by skill other than luck. |
|---|