@article {Francq:2013:1369-7412:345,title = "Optimal predictions of powers of conditionally heteroscedastic processes",
journal = "Journal of the Royal Statistical Society: Series B (Statistical Methodology)",
parent_itemid = "infobike://bpl/rssb",
publishercode ="bp",
year = "2013",
volume = "75",
number = "2",
publication date ="2013-03-01T00:00:00",
pages = "345-367",
itemtype = "ARTICLE",
issn = "1369-7412",
eissn = "1467-9868",
url = "http://www.ingentaconnect.com/content/bpl/rssb/2013/00000075/00000002/art00007",
doi = "doi:10.1111/j.1467-9868.2012.01045.x"
abstract = "
Summary. In conditionally heteroscedastic models, the optimal prediction of powers, or logarithms, of the absolute value has a simple expression in terms of the volatility and an expectation involving the independent process. A natural procedure for estimating this prediction
is to estimate the volatility in the first step, for instance by Gaussian quasimaximumlikelihood or by least absolute deviations, and to use empirical means based on rescaled innovations to estimate the expectation in the second step. The paper proposes an alternative onestep
procedure, based on an appropriate nonGaussian quasimaximumlikelihood estimator, and establishes the asymptotic properties of the two approaches. Asymptotic comparisons and numerical experiments show that the differences in accuracy can be important, depending on the
prediction problem and the innovations distribution. An application to indices of major stock exchanges is given.",
author = "Francq, Christian and Zako{\"ı}an, JeanMichel",
}