In the past, little research has been done about insiders and companies with control. This thesis studies the trading performances of insiders and relational companies. This thesis observed investment company patterns, divided into companies with control and companies without control. We have researched companies with control because we thought that companies with control have characters similar to insiders. We examine this argument directly in two parts. In the first part, we estimate insiders and companies with control to examine whether have the same CSAR by t-test. In the second part, we use the multiple regression method to try to explain what factors a ect CSAR and the factors of companies or insiders that would affect the CSAR. Generally speaking, we nd evidence consistent with the theory that companies with control rights have the same performance as insiders.