Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We develop asymptotic theory for linear statistics of sample autocorrelations of squared residuals from an ARCH(p) model. Our results are valid for most estimators used in practice and do not depend on the assumption of normality of the errors; the existence of the fourth moment is required. In several special cases, we identify the limit distributions as well-known functionals of a Brownian motion.