An application of support vector machines in the prediction of acquisition targets: evidence from the EU banking sector

F. Pasiouras, C. Gaganis, Sailesh Tanna, C. Zopounidis

    Research output: Chapter in Book/Report/Conference proceedingChapter

    Abstract

    In this paper, we investigate the relative performance of both linear and non-linear support vector machines (SVM) models with a polynomial and an RBF kernel in the development of classification models for the prediction of EU bank acquisition targets. The training sample consists of 274 banks, half of which were acquired between 1998 and 2001. The validation sample consists of 31 banks acquired during 2002, and 429 non-acquired banks. We use eight financial variables reflecting the following bank characteristics: capital strength, profitability, efficiency in expenses management, loan activity, liquidity, size, growth, and market power. The models are evaluated in terms of their classification accuracy, as well as with ROC analysis. In both cases, the differences among the models are only marginal.
    Original languageEnglish
    Title of host publicationHandbook of Financial Engineering
    EditorsC. Zopounidis, M. Doumpos, P.M. Pardalos
    PublisherSpringer
    Pages431-456
    VolumeIV
    ISBN (Print)978-0-387-76681-2, 978-0-387-76682-9
    DOIs
    Publication statusPublished - 2008

      Fingerprint

    Bibliographical note

    The full text of this item is not available from the repository.

    Keywords

    • acquisitions
    • banking
    • classification
    • support vector machines

    Cite this

    Pasiouras, F., Gaganis, C., Tanna, S., & Zopounidis, C. (2008). An application of support vector machines in the prediction of acquisition targets: evidence from the EU banking sector. In C. Zopounidis, M. Doumpos, & P. M. Pardalos (Eds.), Handbook of Financial Engineering (Vol. IV, pp. 431-456). Springer. https://doi.org/10.1007/978-0-387-76682-9_14