LARGE SAMPLE DISTRIBUTION OF WEIGHTED SUMS OF ARCH(p) SQUARED RESIDUAL CORRELATIONS
研究了ARCH(p)模型平方残差样本自相关的线性统计量的渐近理论，结果适用于大多数实际估计量，不依赖误差正态性假设，但要求四阶矩存在。
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.