Distribution-Free Tests for Time Series Models Specification Articles uri icon

publication date

  • April 2010

start page

  • 128

end page

  • 137

issue

  • 2

volume

  • 155

International Standard Serial Number (ISSN)

  • 0304-4076

Electronic International Standard Serial Number (EISSN)

  • 1872-6895

abstract

  • We consider a class of time series specification tests based on quadratic forms of weighted sums of residuals autocorrelations. Asymptotically distribution-free tests in the presence of estimated
    parameters are obtained by suitably transforming the weights, which can
    be optimally chosen to maximize the power function when testing in the
    direction of local alternatives. We discuss in detail an asymptotically
    optimal distribution-free alternative to the popular Box-Pierce when
    testing in the direction of AR or MA alternatives. The performance of
    the test with small samples is studied by means of a Monte Carlo
    experiment.