Research On Strategy Optimization Of Weekly Momentum Based On Trading Volume And Non-system Risk

Researching on momentum phenomenon has been for many years. It was found that by taking the momentum (Buy winners and Sell losers),inventors can get a larger stock excess returns. The trading volume contains a lot of information, some scholars combined volume with medium and long term price momentum. Constructing investment strategy of buying low volume winners and selling high trading volume loser can bring greater rewards. However, researches on momentum effect with trading volume on short-term (within one week) are very little, and now the majority of fund managers focus on short-term stock investment, it has a high practical significance for the study of short-term stock momentum.To explore the trading volume and non system risk of momentum investment strategy optimization. Firstly, focus on the American market research, selecting all listed companies stocks of the Nasdaq, the NYSE and AMEX during1965 to 2012;for the Chinese market research, mainly use Shanghai and Shenzhen A-share data during 1993 to 2014. Firstly research on the traditional momentum portfolios, secondly in traditional momentum strategies with trading volume and non system risk constructing investment strategy of buying low trading volume winner and selling high volume loser and buying low non system risk winner and selling high non systematic risk loser. Comparing with traditional week momentum, we find that weekly momentum is existed in the United States market. Considering adding trading volume and non system risk can not only increase returns, persistent of investment positive returns at least 4 years,more than traditional momentum. Chinese market share the same result. It still exist weekly momentum effect, and added trading volume can also improve the stock returns and the persistent of positive returns can reach 5 years; adding non systemic risk is the same. Through regression analysis on weekly momentum strategy based on trading volume and market state, found that has a reverse relation with market volatility and market non liquidity. Different from previous studies, sample after 2000, the effect still exists, based on four factor model of non system risk can still produce the same results.