Longzhen Fan (School of Management, Fudan University)

Title: Predictability of Stock Returns in China Stock Market

Abstract: Predictability of stock returns in China stock market with past 
information is reported. First, we give a brief description of China stock 
market, including both the Shanghai Stock Exchange and the Shenzhen Stock 
Exchange, and then study the predictability of stock returns with different 
information resources. One part of the information is market data, 
including price, trading volume, total market value, and A-shares market 
value of a listing firm. The other part of information is fundamental data, 
including book values, earnings etc. Following results are reported:

(1)  Random walk is a good pattern to describe individual stock prices or 
     stock indexes. 
(2)  Today's total market value, ratio of total market value to book market
     value, ratio of A-shares market value to total market value, ratio of 
     price to earning etc can be used to predict next month's stock returns 
     obviously.
(3)  Predictability of next month's return with above factors has close
     interaction. Three factors can be constructed to absorb all these 
     factors's predictability.
(4)  Three-factor model can be used to explain the indexes's different
     behaviors, and to judge mutual fund.
(5)  Finally, the implications of predictability on passive or active
     investment strategies are discussed.