Sparse-patterned wavelet neural networks and their applications to stock market forecasting

Jack Penm, R. D. Terrell

    Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review


    Wavelet neural networks combine the theories of wavelet analysis and neural networks. This Chapter proposes construction approaches to develop sparse-patterned wavelet neural networks, which demonstrate the ‘presence and absence’ restrictions on the coefficients of a subset time-series system. To demonstrate the effectiveness of the proposed nonlinear approaches, the developed sparse-patterned wavelet neural networks are applied to stock market forecasting.

    Original languageEnglish
    Title of host publicationNonlinear Financial Econometrics
    Subtitle of host publicationMarkov Switching Models, Persistence and Nonlinear Cointegration
    PublisherPalgrave Macmillan
    Number of pages10
    ISBN (Electronic)9780230295216
    ISBN (Print)9780230283640
    Publication statusPublished - 1 Jan 2010


    Dive into the research topics of 'Sparse-patterned wavelet neural networks and their applications to stock market forecasting'. Together they form a unique fingerprint.

    Cite this