augmented arch models for financial time series stability：金融时间序列稳定性的增强模型精品.pdfVIP

Augmented ARCH Models for Financial Time Series: Stability conditions and empirical evidence Robert M. Kunst Johannes Kepler University A-4040 Linz-Auhof, Austria and Institute for Advanced Studies Stumpergasse 56, A-1060 Vienna, Austria Applied Financial Economics, 1997, 7, 575–586 Abstract The class of conditionally heteroskedastic models known as ‘aug- mented ARCH’ encompasses most linear ‘ARCH’-type models found in the literature and, in particular, two basic ARCH variants for auto- correlated series: Engle (1982) explains conditional variance by lagged errors, Weiss (1984) also by lagged observations. The framework per- mits an evaluation of whether the restrictions evolving from the Engle or the Weiss models are valid in practice. Time series of stock market indexes for some major stock exchanges yield empirical examples. In most cases, the statistical approximation to actual dynamic behavior is improved substantially by considering augmented ARCH structures. I INTRODUCTION For some time now, scientific interest in serially correlated volatility has been soaring. This interest is concentrating primarily on financial time series where prediction of means is notoriously unrewarding and hence structure, if any, is to be found through higher-moments proper- ties only. For many financial price variables, including common stocks 1 and stock market indicators, the theory of efficient markets implies that (logarithms of) time series follow random walks. Hence, their first differences are unpredictable while forecasts on the se