Research On Decision-making Quantifying Of Algorithmic Trading In Chinese Stock Market Based On High-frequency Data

Algorithmic trading is the use of computer programs for entering trading orders with the computer algorithm deciding on aspects of the order such as the timing, price, or quantity of the order, or in many cases initiating the order without human intervention.When institutional investor is involved with large transaction volumn, he must consider the execute costing besides commission and transaction tax:because of the limited liquidity,investors will confront impact cost if they want to execute the volumn at once which will make the price change unexpectedly.but if they divide the orders too tiny, the transaction time will increase. More brokers and institutional investors prefer algorithmic trading because of the problem. Today over 90 percents security managers will use algorithmic trading when building investment porfolio.and 40 percents of the trades is finished by algorithmic trading.Algorithmic trading usually have to execute trade orders frequently and follow the real time security market situation.so low frequency trading data is far from enough for this king of researches.building the model with high frequency trading data will give better fit result.Algorithmic trading in china stock market is at starting stage,chinese stock market have a bigger price volatility and liquidity,so algorithmic trading is more valuable. As the stock index and futures is raised, institutional investor is focused on the design of arbitrage model, they need algorithmic trading to help them avoid the volatility of market price urgently.We first introduce the development of algorithmic trading and its design philosophy ,then give some advice on the execution of algorithmic trading in chinese stock market.We proposed a new trade algorithm based on chinese stock market ,Strategy without interactive effect first build Autoregressive Conditional Duration model to get a predictive transaction duration time series,then predict the volumn distribution and price change to get the adjusted period volumn.besides,we combine interactive factor by event analysis for strategy with interactive effect.We take the emprical research with the high frequency trading data of china stock market of 2008.by building the duration model, volumn model and price prediction model ,we get the executed trade volumn, result indicate that the simulated price is better than the average price by traditional VWAP. Furthermore,we add the interactive effect on the model ,results show that the exrcutive price is lower than price without ineractive effect under the same deal completation constrant.