Empirical Study Of Market Quality And Price Discovery Based On High-frequency Data

Based on high frequency quotation and transaction data,this paper studies the market quality and price discovery process in Chinese stock market,and further explores the contribution of different quotation to price discovery in the order book.Firstly,for the depiction of market quality,this paper mainly uses the realized bipower variation and Glostern-Harris transaction cost model.The results show that the market volatility and transaction costs in June 2015 are higher than those in June 2016,and the liquidity is relatively poor,that is,the market quality is relatively poor in June 2015.In addition,turnover rate,offer rate and trading frequency in June 2015 were also significantly higher than those in June 2016.Secondly,from the perspective of time series analysis,this paper discusses the relationship between transaction frequency and market quality.The results show that,in 2016,the trading frequency and volatility are not the Granger Cause of each other,the market depth is the Granger Cause of the trading frequency,the market width and the trading frequency are not the Granger Cause of each other.The Granger Cause is only a relationship in the statistical sense,which does not represent the real cause,so it is only useful for the prediction of variables.In 2015,due to the poor stability of data,the results are of little economic significance.Besides,this paper measures the difference of the property of stocks with different trading frequency based on cross section data.The results show that,with the increase of market value,the market volatility and transaction cost decline.That is,with the increase of market value,the nature of the stock appears to be better.While with the increase of trading frequency,the market volatility and transaction costs rise.That is,the stocks with high trading frequency often exhibit poor nature.Finally,this paper conducts an in-depth research on the order book,and discusses the contribution of different quotes to price discovery process.The variance ratio is used to analyze the market efficiency.The results show that the market efficiency increases with the increase of the trading frequency in some sense.The study shows that the best quotation expressed by the intermediate of best bid and ask has a high contribution to the price discovery,with an average contribution of 69.12%,while the suboptimal offer has only 30.88%.The information share in the optimal quotation is large.And the higher the trading frequency is,the more the information share in the optimal quotation.There is a certain asymmetry in the contribution of bid and ask to the price discovery,and the asymmetry is more common in the stocks with small market value.What's more,high frequency trading is usually accompanied by ask orders,which makes the order book contain more information.In general,although the cross section data shows that high frequency trading makes the market quality decline,the increase of trading frequency also increases the efficiency of the market.Therefore,we should use high frequency trading reasonably and effectively to better avoid the risk and improve the quality of the market.