Statistical inference for time-varying ARCH processes

In this paper the class of ARCH(infinity) models is generalized to the nonstationary class of ARCH(infinity) models with time-varying coefficients. For fixed time points, a stationary approximation is given leading to the notation "locally stationary ARCH(infinity) process." The asymptotic properties of weighted quasi-likelihood estimators of time-varying ARCH(p) processes (p < infinity) are studied, including asymptotic normality. In particular, the extra bias due to nonstationarity of the process is investigated. Moreover, a Taylor expansion of the nonstationary ARCH process in terms of stationary processes is given and it is proved that the time-varying ARCH process can be written as a time-varying Volterra series.