A Study On The Informed Trading And Strategic Behavior In An Order-Driven Market

The securities markets are quite different from the commodities markets. Scattered and asymmetrical information are their salient features. Different traders, including informed traders and uninformed traders, their market behavior interact and affect each other, and thus it makes the price disclose private information gradually.This dissertation firstly construct an order-driven market probability model of informed trading to analyze the correlation between informed trade and price variation and the return of assets, explore the strategic private information utilizing behavior of informed traders during a week at the same time. Secondly, this dissertation focus on the informed trading based on marketwide private information, study the correlation between the industry-wide private information and stock block's comovement, and the probability of informed trading, analyzing how macroeconomic announcement affect the information asymmetry in the stock market. Thirdly, the dissertation test whether trade duration or price duration contains private information or not, test the stealth trading hypothesis. Finally, we set up a sequential trading model of order-driven market to study the factors that affect the trader's order selection between market order and limit order, and the trader's order selecting strategy.When measuing the probability of informed trading based on firm specific private information, this paper finds:first, the return pattern of stock may be devided into two states, one is the liquid trading state without important information enterning into the market; another is the information trading state with new information incorporating into the stock price, the former state has lower return means and variance, while the later has larger return means and variance. Informed traders'payoffs are the losses of uninformed traders. Second, after using the time series of weekly probabilities of informed trading to deduce the strategic private utilizing behavior of informed traders, the result indicates the informed traders choose to trade mainly by using public information at the beginning of a week, and then use the private information after the beginning. Third, there exsits positive correlation between informed trading and price variation, the probability of informed trading can explain asset return. The larger the information risk, the higher the excess return of stock return.Regarding the correlation between market wide private information and informed trading, this paper finds that when the industry-wide-information order flow increase, the uninformed traders would be more sure of their inferring about the market condition, and the probability of informed trading will decrease. And the empirical analysis also shows that industy-wide private information affects all the firms belonging to the same industry at the same direction. Moreover, if traders own heterogeneous information about macroeconomic variables, and the heterogeneous information are not yet aggeragated by the market; or the markrt had responded to some macroeconomic information, but the information's announcement have lagging; or insider traders have insider information about macroeconmic news, the information asymmetry will increase during the announment period of macroeconomic information. This paper points the possibilities of the last circumstances using the stamp duty adjustment as an example.Regarding the investigation of whether the trade duration (price duration) contains private information, this dissertation use the asymmetric autoregression conditional marked duration model to test the negative correlaton between price duration and the substitute variables of information asymmetry. The empirical result shows the higher the information asymmetry, the shorterer the price duration. The test on the stealth trading shows that informed traders don't adopt the strategy of slicing the orders so to hide the private information in china stock market owing to the higher transaction costs.In the analysis of order submission strategy of informed traders and uninformed traders, this dissertation shows that trader's evaluation on the assets and the private information, the probability of submitting market order, the ratio of informed traders, the ratio of traders who have higher evaluation about the assets and the tick size affect the order selection of informed traders and uninformed traders. Generally, informed traders prefer market order in order to reach a deal quickly and obtain the information payoffs in time, while uninformed traders prefer limit order so as to obtain the price impovement.Moreover, the main contributions of the dissertation are indicated as follows. First, this dissertation clearly shows the suitability of trade direction algorithms in china stock trading datasets and the details of modeling trade durations. Secondly, this dissertation tests the explainnation power of the probability of informed trading on the monthly stock return. Thirdly, the dissertation studies the stock block comovement using rhe industry-wide private information. Forthly, the dissertation deduces the inter-temporal behavior pattern of informed traders during a week, and tries to prove that the informed traders prefer to submit market order and the uninformed traders prefer to submit limit order.