Advanced Search
MyIDEAS: Login

Modelo no lineal basado en redes neuronales de unidades producto para clasificación. Una aplicación a la determinación del riesgo en tarjetas de crédito = Non-linear model for classification based on product-unit neural networks. An application to determine credit card risk

Contents:

Author Info

  • Martínez Estudillo, Francisco José

    ()
    (Departamento de Gestión y Métodos Cuantitativos, ETEA Córdoba (España))

  • Hervás Martínez, César

    ()
    (Departamento de Informática y Análisis Numérico, Universidad de Córdoba)

  • Torres Jiménez, Mariano

    ()
    (Departamento de Gestión y Métodos Cuantitativos, ETEA Córdoba (España))

  • Martínez Estudillo, Andrés Carlos

    ()
    (Departamento de Gestión y Métodos Cuantitativos, ETEA Córdoba (España))

Registered author(s):

    Abstract

    El principal objetivo de este trabajo es mostrar un tipo de redes neuronales denominadas redes neuronales basadas en unidades producto (RNUP) como un modelo no lineal que puede ser utilizado para la resolución de problemas de clasificación en aprendizaje. Proponemos un método evolutivo en el que simultáneamente se diseña la estructura de la red y se calculan los correspondientes pesos. La metodología que presentamos se basa, por tanto, en la combinación del modelo no lineal RNUP y del algoritmo evolutivo; se aplica a la resolución de un problema de clasificación de índole económica, surgido del mundo de las finanzas. Para evaluar el rendimiento de los modelos de clasificación obtenidos, comparamos nuestra propuesta con varias técnicas clásicas, como la regresión logística o el análisis discriminante, y con el clásico modelo de perceptrón multicapa de redes neuronales basado en unidades sigmoides y el algoritmo de aprendizaje de retropropagación (MLPBP) = The main aim of this work is to show a neural network model called product unit neural network (PUNN), which is a non-linear model to solve classification problems. We propose an evolutionary algorithm to simultaneously design the topology of the network and estimate its corresponding weights. The methodology proposed combines a non-linear model and an evolutionary algorithm and it is applied to solve a real economic problem that occurs in the financial management. To evaluate the performance of the classification models obtained, we compare our approach with several classic statistical techniques such us logistic regression and linear discriminat analysis, and with the multilayer perceptron neural network model based on sigmoidal units trained by means of Back-Propagation algorithm (MLPBP).

    Download Info

    If you experience problems downloading a file, check if you have the proper application to view it first. In case of further problems read the IDEAS help page. Note that these files are not on the IDEAS site. Please be patient as the files may be large.
    File URL: http://www.upo.es/RevMetCuant/art10.pdf
    Download Restriction: no

    File URL: http://www.upo.es/RevMetCuant/art10.txt
    Download Restriction: no

    Bibliographic Info

    Article provided by Universidad Pablo de Olavide, Department of Quantitative Methods for Economics and Business Administration in its journal Revista de Métodos Cuantitativos para la Economía y la Empresa = Journal of Quantitative Methods for Economics and Business Administration.

    Volume (Year): 3 (2007)
    Issue (Month): 1 (June)
    Pages: 40-62

    as in new window
    Handle: RePEc:pab:rmcpee:v:3:y:2007:i:1:p:40-62

    Contact details of provider:
    Postal: Carretera de Utrera km.1, 41013 Sevilla
    Phone: + 34 954 34 8913
    Fax: + 34 954 34 9339
    Email:
    Web page: http://www.upo.es/economia/metodos/
    More information through EDIRC

    Related research

    Keywords: clasificación; redes neuronales de unidades producto; redes neuronales evolutivas; classification; product unit neural networks; evolutionary neural networks;

    Find related papers by JEL classification:

    References

    No references listed on IDEAS
    You can help add them by filling out this form.

    Citations

    Lists

    This item is not listed on Wikipedia, on a reading list or among the top items on IDEAS.

    Statistics

    Access and download statistics

    Corrections

    When requesting a correction, please mention this item's handle: RePEc:pab:rmcpee:v:3:y:2007:i:1:p:40-62. See general information about how to correct material in RePEc.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Rocío Fernández).

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If references are entirely missing, you can add them using this form.

    If the full references list an item that is present in RePEc, but the system did not link to it, you can help with this form.

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your profile, as there may be some citations waiting for confirmation.

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.