The Research About Measuring And Pricing Information Risk In Chinese Stock Markets

Measuring time-varying information risk of stocks is of great significance forinvestors, companies and regulators. It contributes to measuring information risk moreaccurately. Therefore, it is not only helpful to controlling information risk forinvestors, managing the capital cost for enterprises and regulating the market forregulators, but also useful for the asset pricing. The measurement and pricing issuesof time-varying information risk of Chinese stock markets are studied theoreticallyand empirically by this dissertation. The dissertation consists of four parts:(1) thebackground, significance, introduction to structure arrangement and the content,innovations of the dissertation and the summary of the relevant research area (chapter1~2);(2) the research of the measurement of time-varying information risk(chapter3~5);(3) the study on discount phenomena of H shares based on the time-varyinginformation risk measurement(chapter6);(4) finally, the conclusions andprospects(Chapter7). The detailed contents are as below:In chapter1, the background and significance, the content, the structure andinnovations of the dissertation are introduced.In chapter2, the summary of measuring information risk relevant researches isreviewed systematically. The deficiency of the present researches and the latestresearch trends are pointed out at the end of this chapter.In chapter3, to overcome the assumption of constant probabilities of the states ofnews and symmetric order-flow shock of the model of Duarte and young (2009), anew time-varying information risk measurement model is proposed which extends themethod of Duarte and young (2009). The probabilities of the States of News and theprobabilities of Symmetric Order-Flow Shock by using trade volume are modeled inthe new model which allows the probabilities of the States of News and theProbabilities of Symmetric Order-Flow Shock both to vary. Then several activelytraded stocks are selected for testing the new model.In chapter4, based on the MMPP model, a new time-varying information riskmeasurement model is put forward. The EKOP model is based on the assumptionsthat the arrival of information is independent on each day and the arrival rates ofinformed and uninformed trade are both constant, which are too strict. The newmethod relaxes these assumptions. First, the number of trades of liquidity trade andinformed trade are modeled, based on the Markov Modulated Poisson Process Model(MMPP) in order to separate the total trades. Then, a new method to estimate PIN is proposed according to this separation model. Finally, the PIN of several activelytraded stocks is estimated, and the appropriateness of our method is gauged.In chapter5, Taking the short sale constraint into account, a new method toestimate PIN based on the study of Tay,Ting,Tse and Warachka（2009）is introduced.Then both intraday and daily PINs of several actively traded stocks are estimated bythe new method. The result of the new method is contrasted to the results ofTay,Ting,Tse and Warachka（2009）.In chapter6, the time-varying information risk of A shares and H shares ismeasured by using the method which is put forward in chapter3, since the existingresearches about the H shares discount mostly measure information risk by indirectmethod. Whether information risk is one cause of the discount of H shares is tested bycontrolling the relative variables.In chapter7, the conclusions and some prospects of the dissertation are made.