Answers are grounded in this paper and cite their sources.
We derive the asymptotic distribution of the sequential empirical of the squared residuals of an ARCH(p) sequence. Unlike the of an ARMA process, these residuals do not behave in this context asymptotically independent random variables, and the asymptotic involves a term depending on the parameters of the model. We show in certain applications, including the detection of changes in the of the unobservable innovations, our result leads to distribution free statistics.
No takes yet. Share an insight, caveat, or question.
Horváth et al. (2001) studied this question.