E-Prints Complutense

A flexible Tool for Model Building: the Relevant Transformation of the Inputs Network Approach

Impacto

Descargas

Último año



Pérez Amaral, Teodosio y Gallo, Giampiero M. y White, Halbert (2002) A flexible Tool for Model Building: the Relevant Transformation of the Inputs Network Approach. [ Documentos de trabajo del Instituto Complutense de Análisis Económico (ICAE); nº 01, 2002, ]

[img]
Vista previa
PDF
117kB

URL Oficial: http://eprints.ucm.es/7650/


URLTipo de URL
https://www.ucm.es/icaeInstitución


Resumen

A new method, called relevant transformation of the inputs network approach (RETINA) is proposed as a tool for model building and selection. It is designed to improve some of the shortcomings of neural networks.
It has the flexibility of neural network models, the concavity of the likelihood in the weights of the usual likelihood models, and the ability to identify a parsimonious set of attributes that are likely to be relevant for predicting out of sample outcomes.
RETINA expands the range of models by considering transformations of the original inputs; splits the sample in three disjoint subsamples, sorts the candidate regressors by a saliency feature, chooses the models in subsample 1, uses subsample 2 for parameter estimation and subsample 3 for cross-validation. It is modular, can be used as a data exploratory tool and is computationally feasible in personal computers.
In tests on simulated data, it achieves high rates of successes when the sample size or the R2 are large enough. As our experiments show, it is superior to alternative procedures such as the non negative garrote and forward and backward stepwise regression.


Tipo de documento:Documento de trabajo o Informe técnico
Palabras clave:RETINA
Materias:Ciencias Sociales > Economía > Comercio
Título de serie o colección:Documentos de trabajo del Instituto Complutense de Análisis Económico (ICAE)
Volumen:2002
Número:01
Código ID:7650
Depositado:26 Feb 2008
Última Modificación:06 Feb 2014 07:55

Descargas en el último año

Sólo personal del repositorio: página de control del artículo