A Research On Price Impact And Intra-day Pattern Of Liquidity In Chinese Stock Market

The fundamental purpose of the stock market is to provide investors with the opportunity to trade stocks and improve the liquidity of the stock. In particular someone says "liquidity is the stock market's vitality." Therefore, the causes and influence factors of liquidity cost are the research emphases of market microstructure theory, it helps to improve the function and efficiency of the stock market.This paper combines theoretical analysis and empirical research methods, from various angles of the stock market's liquidity, including the causes and influence factors of liquidity cost, estimating price impact function, and causes and application of liquidity intraday patterns. This paper also discusses the liquidity based on different measures, including price impact coefficient and the trading volume. There are four main parts of text:Comparatively speaking, as part of the cost of liquidity, adverse selection cost is more endogenous, so this paper chooses to research causes and influence factors of price impact based on the adverse selection in trading at first. This paper borrows ideas from analytical framework of Easley and O'Hara (1987), and receives the liquidity impact factor model based on adverse selection as its application to the order-driven market. Simultaneously this paper tests how various adverse selection variables influence the impact coefficient. The results show that the proportion of the top ten shareholders on behalf of the proportion of informed traders have an obvious effect of impact coefficient, reflecting the recipients'high sensitivity and disgust because of informational disadvantaged from ownership concentration. Only volatility of the historical returns on small cap stocks is significant, indicating that the difference of available information from different size of capital stock influences level of liquidity, that is the difference between the level of adverse selection causes differences in liquidity. The probability of information events' occurrence is also significant, indicating traders are worried the reveal of report in advance. The price responds to the expected results of different events differently. Liquidity impact coefficient is negatively correlated with expected stock return, that is good expectation increases levels of liquidity, but bad expectation lowers levels of liquidity when adverse selection cost becomes very high. For the Chinese stock market, our conclusions show that we can improve the level of stock market's liquidity by reducing the information asymmetry between traders. In addition, this study will also help investors to more accurately estimate the impact coefficient and establish the trading strategies.We study the difference of liquidity impact between time and companies based on adverse selection. The price impact function discusses the level of impact under a certain volume directly, trying to find the function relation between volume and price changes in a moment. Based on previous studies, the second part of this paper concentrates on the function's form and the selection of variables. Empirical studies indicate that the price impact function is concave power function or its linear combination in Chinese stock market. The most appropriate variables of this function are return and trading volume standardized by a 30-day moving average volatility and a 30-day moving average trading volume. Study also shows that the impact of large shares is smaller than small companies by group tests, and the price impact of the buy and sale order does not appear a significant difference.In the real market, we call the adverse change in transaction price because of lack of liquidity as execution cost, which is not only prevalent in the market, and is a major component of transaction costs of the institutional investors. In view of this, institutional investors always slice large amount and trade them in several days, and carry out several transactions every day, in order that the market can gradually digest orders, thereby reducing the execution cost. In theory, the infinite number of transactions can make the impact of the transaction negligible. But in this case, the risks exposure of the transaction positions will become great in a long time, which is undesirable for risk-averse investors. In contrast, the other extreme is to complete the transaction instantly, which can reduce the risk exposure to zero. But the execution cost increases rapidly, which is not desirable too. The feasible execution strategy should reach the compromise between execution costs and risks given certain trader risk aversion coefficients. The third part of the paper follows the framework of Almgren and Chriss (2000) to explain the intra-day's U-shaped liquidity mode based on trading volume.A large number of empirical studies have shown that when opening the stock volume is relatively large, then gradually decreased, until near the close up again. Throughout the whole day, trading volumes at the opening and closing are significantly larger than in other periods, this distribution is also known as intra-day's U-shaped liquidity mode or trading volume mode, and China's stock market is not an exception. Previous researchers explain this phenomenon as uninformed traders' clustering to trade or their portfolio reconstruction at opening and closing. However, since the beginning of this century, the algorithm trading quickly becomes the most dominant trading means, which did not appear formerly. This paper regard it as an important reason causing the intra-day U-shaped trading volume pattern. The third part constructs the intra-day optimal trading strategies under the assumption that representative traders are risk-averse and not fully informed. As the assumption that traders receive new information about the future value of stock before the opening, the characteristics of risk aversion tend to make them complete the transaction as soon as possible. But subject to liquidity impact, the transaction orders have to be split. As trade-off between risk exposure and liquidity impact, the optimal allocation strategy is the distribution that execution amount gradually decreases with time after the opening, which leads to high trading volume when opening. This feature of not fully informed makes traders observe the market price changes continuously and update their expectations of the stock value and the total number of transactions. This means that it is closing at that the release of information is the most adequate in the day, and the trading volume is high. By two points above, we have a new perspective to explain the U-shaped intraday trading volume mode.Since the intra-days U-shaped trading volume pattern is decided by the trading behavior of investors endogenously. if some trading volume in a moment does not meet the law, according to the classical relation between price and trading volume, the non-expected trading volume contains information on the stock's future return. The last part of paper explores this type of relationship between price and trading volume. The difference from the previous researchers is that calculation about non-expected trading volume is based on the intra-days U-shaped volume pattern, which tests the short-term predictability of the trading volume on the return. The empirical results based on high-frequency transaction data show that, in all four cases, the non-expected trading volume that trading volume reduce and price rise has the best predictability, which almost means the subsequent rise in that day. We also test the profitability of this trading strategy and the answer is yes in the case of without regard to transaction costs. Especially, using the seven minutes accumulated returns and trading volume, the strategy that buying stock index when trading volume reduce and price rise then selling all position 180 minutes later can gain 47.1% of the annual rate of return.According to the conclusions of this paper, there exist several drawbacks about the liquidity of Chinese stock market, including the lack of information disclosure, insider trading, price manipulation and high transaction costs. The last chapter of the paper proposes construction and regulatory policy proposals of Chinese stock markets in accordance with but not limited to these issues, including improving information disclosure, strengthening supervision on illegal transactions, the gradual introduction of market maker, improving the block trading system and reducing transaction costs. We believe the implementation of these policy measures will greatly improve the liquidity in Chinese stock market.