Portfolio selection problem using linear and gaussian graphical models with penalization
Many alternatives have been proposed to improve the classic Markowitz mean-variance approach in the portfolio selection problem. Some alternatives focus in decreasing the error of the estimations of the expected return vector and the covariance matrix. Following this framework, in this paper we are...
- Autores:
-
Lloreda Palacios, Ricardo Andrés
- Tipo de recurso:
- Trabajo de grado de pregrado
- Fecha de publicación:
- 2017
- Institución:
- Universidad de los Andes
- Repositorio:
- Séneca: repositorio Uniandes
- Idioma:
- eng
- OAI Identifier:
- oai:repositorio.uniandes.edu.co:1992/39749
- Acceso en línea:
- http://hdl.handle.net/1992/39749
- Palabra clave:
- Administración del portafolio
Portafolio de inversiones
R (Lenguaje de programación de computadores)
Ingeniería
- Rights
- openAccess
- License
- https://repositorio.uniandes.edu.co/static/pdf/aceptacion_uso_es.pdf
| Summary: | Many alternatives have been proposed to improve the classic Markowitz mean-variance approach in the portfolio selection problem. Some alternatives focus in decreasing the error of the estimations of the expected return vector and the covariance matrix. Following this framework, in this paper we are going to address this problem with penalizations over the asset's weights estimations, in both linear and graphical models. With these penalizations, we induce two important properties to the portfolios obtained, sparsity and stability. These properties yield a better out-of-sample performance for a portfolio because they decrease the error carried by the parameters estimations and also reduce the number of assets in a portfolio. Thus, they are desired for an investor |
|---|