Abstract

GARCIA SALGADO, Oswaldo  and  MORALES CASTRO, Arturo. FINANCIAL PERFORMANCE OF FIRMS: A PROPOSAL FOR CLASSIFICATION RNA. Dimens.empres. [online]. 2016, vol.14, n.2, pp.11-23. ISSN 1692-8563.  https://doi.org/10.15665/rde.v14i2.686.

By using artificial neural networks [ANN] it was proposed to improve the accuracy of classification of companies within the Bolsa Mexicana de Valores [BMV], specifically the commercial sector compared to multiple discriminant analysis techniques [MDA] and logit models. More than fifty neural architectures were developed, and artificial neural network that resulted was the MLP architecture 6: 12: 2 based learning algorithms retro-backpropagation. The results found with the technique [ANN] showed that this technique has a better prognosis assessment and classification obtained by ADM and logit techniques.

Keywords : Financial Markets; Financial Performance; logit model; ADM Techniques; Artificial Neural Networks.

        · abstract in Spanish | Portuguese     · text in Spanish     · Spanish ( pdf )