It is a common feature in the capital market that money managers who possess professional skill make decisions on behalf of their principals. The money managers gradually become the main investors, and play a significant role in the market. Since the money managers’ actions are assumed to be unobservable on the part of the principals and consequently cannot be contracted upon there is room for moral hazard. An extensive literature on money managerial compensation contracts has demonstrated that to mitigate the moral hazard problem the agents’compensation should be linked to principals’ wealth. Equity-linked compensation is a natural way to achieve this end. While these economic theories on resolving the principal-agent problem are fairly well developed at this point, most of them assume that people behave rationally and optimally. It is puzzled in adopting the policy implications of such theories because they will change if more realistic assumptions were made about human behavior. The spring up and development of behavioral finance have turned the scale that assuming people are rational and optimal, and it advances that the investors just have bounded rationality. Therefore, this thesis explores motivation problems of money managers in capital market based on bounded rationality, under the condition that existing literatures on behavioral finance pay much attention on the systemic effects of investors’ irrational behaviors on capital market, and less on money manager’s bounded rationality, and the theory of motivation emphasizes physical motivation. The purpose of this dissertation is to provide new ideas for money managers’ motivation from behavior finance angle of view.The main research of this dissertation is as follows:(1) The design of principal-agent contract based on investing sentiment. By combining the behavioral finance theory with the principal-agent theory, this dissertation advances that it is necessary to introduce the managers’ investing sentiment into the principal-agent problem in the capital market. By establishing the mathematical model, this dissertation analyzes how the irrational money managers’ investment decisions affect the principals, and it derives the optimal contract between the principals and the money managers. The result shows that the optimistic (pessimistic) sentiment of money manager could compensate principal’s agency cost, and the optimal incentive intensity is decreasing in the optimistic(pessimistic) extent. It indicates that the optimistic (pessimistic) sentiment of money manager is beneficial to the principal under some conditions.(2) Study of static and dynamic principal-agent problem based on overconfidence. First, this dissertation advances to introducing overconfidence into incentive problem and the internal/external monitor mechanism. The result shows that from the micrpcosmic point overconfidence of the money managers could augment the incentive intensity and reduce the agency costs; From the macroscopical point,the proportion of overconfidence money managers that engaged in the market will increase by their average confidence of the market. Second, it derives the dynamic optimal contract based on the alteration of money managers’overconfidence. The result shows that both overconfidence and supervise system make the incentive contracts become cheap, and reduce the agency costs. They are substitutes for the incentive contract in coping with the principal-agent problem. But overconfidence of the manager will decline gradually under some conditions, and the optimal incentives will get back to the level for rational managers finally.(3) Study on team coordination with irrational expectation. Applying the behavioral finance theory to resolving the team coordination problem, this dissertation researches the team members’ behavior, team value and supervision, and analyses the dynamic effect of irrational expectation on team coordination. The result shows that members with excessive expectation will aggravate team’s moral hazard, and reduce its performance, so it is better to monitor them intensively; Contrarily, members with conservative expectation will alleviate team’s moral hazard and upgrade its value, so it is better to monitor them slightly. Moreover, along with the enrichment of working experience, team members will become rational, and the effect of irrational expectation on team coordination will get smaller and smaller.(4) Empirical study on money managers’ bounded rationality in Chinese capital market. By establishing the sentiment index of money managers, this dissertation did the empirical study about whether the money managers in our country behave bounded rationally. Moreover, we studied the relationship between the money managers’ sentiment and the main index reflecting market operation. The results show that, on one hand, the money managers in our contury do have cognize bias about the market, and they are not rational investors; On the other hand, the main indices of the market could reflect the money manager’s sentiment primely, so both the principal and the government could estimate and control the money manager’s behavior by obversing the variety of these indices.At the end of the dissertation, it summarizes the results and contributions of the research, and points out the further research direction in the future.