GAMSEL methodology application to provide a balance between interpretability and flexibility in the bankruptcy prediction problem
"This study compares the prediction accuracy and model interpretability, between a generalized additive model with an incorporated variable selection procedure (GAMSEL), and some representative methodologies used for bankruptcy prediction. Although the most flexible methods, as support vector m...
- Autores:
-
Ramírez Zuluaga, Juan Alberto
- Tipo de recurso:
- Fecha de publicación:
- 2017
- Institución:
- Universidad de los Andes
- Repositorio:
- Séneca: repositorio Uniandes
- Idioma:
- spa
eng
- OAI Identifier:
- oai:repositorio.uniandes.edu.co:1992/61106
- Acceso en línea:
- http://hdl.handle.net/1992/61106
- Palabra clave:
- Administración de riesgos
Quiebra
Redes neurales (Computadores)
- Rights
- openAccess
- License
- https://repositorio.uniandes.edu.co/static/pdf/aceptacion_uso_es.pdf
| Summary: | "This study compares the prediction accuracy and model interpretability, between a generalized additive model with an incorporated variable selection procedure (GAMSEL), and some representative methodologies used for bankruptcy prediction. Although the most flexible methods, as support vector machine SVM) with radial kernel or boosting perform well in this problem givmg high accuracies, they are harder to Interpret". -- Tomado del resumen. |
|---|