| Futures market is an important part of the financial markets as well as the national economy. With its two important functions, price discovery and hedging, futures market has a huge impact on virtual economy, stock market, foreign exchange market, etc. It also has a great effect on the operation of substantial economy for its close connection with spot trading.With the rapid development of international trade and economic integration in recent years, Futures market arises its role for allocating resources across borders and snatching the international pricing power of resources. Therefore, it is important to understand the microstructure of the futures markets and how it works to affect the price formation as well as the price volatility.Market microstructure theory has provided us a great way to understand futures market microstructure and its dynamics. In1995, Maureen O’Hara defined the market microstructure as the study of the process and outcomes of exchanging assets under a specific set of rules. Based on specific trading rules, microstructure theory focuses on how specific market microstructure affects the mechanism of price formation.In2000, Ananth Madhavan surveyed the theory, experimental and empirical study of market microstructure, in which he described market microstructure as an area of finance that studies the process by which investors’ latent demands are ultimately translated into prices and volumes. One of the features of market microstructure study is focusing on the trading mechanism itself to analyze its impact to price formation。And that’s the very background microstructure noise has been proposed and developed. According to the market microstructure theory, market is different from the hypothesized perfect market in real life. Information is asymmetric in reality. There are transaction costs as well as other market frictions. The prices of securities will deviate from the equilibrium price of the assets under the imagined perfect market in the transaction process because of market frictions.We summarize the market frictions caused the deviation from the equilibrium price in trading process as market microstructure noise. It is inherent in the transactions and captures a variety of frictions as bid-ask bounces, discreteness of price changes, differences in trading size, informational content of price changes, gradual response to block trades, strategic component of the order flow, inventory control effects, etc.The emergence of high-frequency data and ultra-high-frequency (UHF) data recorded each pieces of market transactions, has offered important raw materials for the study of futures market microstructure noise.Traditional ways of study on the asset price fluctuations are based on low frequency data, which sampling frequency is the lower and does not contain every transaction. Because of the loss of trading information, it is inappropriate and inadequate to describe the accurate market characteristics.High-frequency data and UHF data including all transaction information would perform better than lower sampling data. However, the high frequency data and UHF have some obstacles to be applied in traditional ways of volatility estimation due to its high sampling frequency and other characteristics. For example, ranging intervals, discrete changes of the price, the periodic mode of the transaction, multiple transactions happens in one second. The usual ways of volatility estimators carried out in various GARCH model or SV(Stochastic Volatility) model or RV(Realized Volatility) model under this situation are no longer asymptotic to integrated volatility. As a matter of fact, the higher the sampling frequency, the more biased are the traditional estimators, which is caused by the contamination of the market microstructure noise.Zhang, et al (2005)proposed two-scale realized volatility (TSRV) to take advantage of the full sample observation data to estimate the real fluctuations of the price in trading process. It is available for the TSRV estimator to converge to the integrated volatility, which offers us a good way to decompose the fluctuations of the price into a fundamental volatility and a market microstructure noise.Based on the TSRV model, this paper makes use of500ms high frequency raw data and processes it into UHF data, to analyze the market microstructure noise in domestic futures market for the first time.And the contents are organized as follows. In chapter1introduced the background of the study, focused questions and author’s arrangements. In chapter2reviewed the theory, experimental and empirical study of the market microstructure noise. In chapter3studied the development and microstructure of domestic futures market. In chapter4described the data and analyzed the periodic mode of inter-day as well as intraday transactions. In chapter5introduced the ways used to estimate microstructure noise, we also used UHF data separated noise and fundamental volatility from observed price based on the TSRV model. In chapter6explained the marker microstructure noise by a simple OLS regression model contains spread, transaction size, TSRV, trading frequency as explanatory variables. Chapter7concludes, followed the paper’s reference and acknowledgements. There are some pictures in the appendix which put for the convenience. |
PDF Full Text Request