上证股市收益长期记忆_基于V_S经验分析_何兴强.pdfVIP

200612 12 : 1000-6788(2006) 12-0047-08 :VPS 何兴强, 李仲飞 ( , 510275) : VPS , AB ,ICSS ,.: AB ,B ;AB ;A , B. 10, 1 ,B A . : ; ;(VPS); (ICSS) : F830191 : A Long-Ter Me ory in Stock Returns of ShanghaiStock Exchange: Evidencefro VPS Statistic HE Xing-qiang, LI Zhong-fei (Lingnan College, Sun Ya-t sen University, Guangzhou 510275, China) Abstract: For the first ti e, the rescaled variance test is applied to investigate the long-ter e ory effect in China. s stock returns. We exa ine thefull sa ple daily stock arket returns of ShanghaiA and B shares. Based on the detection of changes of variance using the iterated cu ulative su of squares algorith , we study the long-ter e ory effect of stock arket returns in different sub-periods. So e rando ly selected stocks are also considered. Results obtained include: there exists little evidence of long-ter e ory in the full sa ple stock arket returns of Shanghai A and B shares, with regard to A shares, the B shares shows relatively ore significant long-ter e ory; there are 2 and 4 notable variance changes in A and B shares respectively, andfor each sub-period of A shares, there does not exist notable long-ter e ory, however, there does exist considerable long-ter e ory in so e sub- periods of B shares. Study of therando ly selected stocks concludesthat a ong the 10selected stocks, only 1stock. s return series displays significant long-ter e ory, andwith regard to A shares, the B shares shows relatively ore significant long-ter e ory. Key words: stock arket; long-ter e ory; rescaled variance (VPS) ; iterated cu u