An Activism Effect Analysis Of Large Shareholders Based On Investment Portfolio Management

The study of large shareholder is an important point in corporate governance research. The share centralized degree of Chinese stock company is very high and the control capability of large shareholder is strong, it is common large shareholder control the stock company. So studying the activism standard of large shareholder is useful to analysis and forecast large shareholder activism in future.In the second chapter, we summarize investigation theories on the large shareholder activism and the activism effect on his company of our country and overseas. Last research found that the share hold proportion, the stock kind and the share hold structure produce influence on the company performance. The aim of large shareholder activism is get benefit which is called control benefit and is divided into mutual benefit and private benefit. The large shareholder activism is divided into misappropriating activism and sustaining activism. This chapter also summarizes the investment portfolio management theory. We think investors should choose such investment portfolio which has highest return rate when his risk is fixed.In the third chapter, we discuss the main study object and four study methods. The first is case analysis method. We show a real case in which large shareholder act base on investment portfolio management. Then we study the effect of large shareholder activism on stock company with event study method. The third is mathematics consequence method. We get five deduct with mathematic model. The last is multivariate linear regression analysis method. We prove the deduction with multivariate linear regression analysis.In the fourth chapter, we discuss the case of GuiLinJiQi company. We found the large shareholder of GLJQ adjust investment proportion between GLJQ and other subsidiary. Some of these adjust activisms steal GLJQ benefit and some activisms sustain GLJQ development. We found these adjust activisms change GLJQ risk level.In the fifth chapter, we found a mathematic model and get five deduct. We first study the return and risk of large shareholder's investment portfolio. And deduct large shareholder five activisms base on fixed risk and highest return.In the sixth chapter, we found multivariate linear regression analysis model base on the five deduct of the fifth chapter. We choose swatch and variable the model need. The result of model calculation shows these deduct mostly accord with realism.This study found that, when two companies, belong to the investment portfolio of large shareholder, have opposite correlation. If one company's risk increased, large shareholder would decrease another company's risk. v.v.This study also found that, when two companies, belong to the investment portfolio of large shareholder, have opposite correlation. If large shareholder increases one company's investment proportion, large shareholder would decrease this company's risk. When two companies, have positive correlation, if one company's risk level was lower than another company's risk multiply two company's correlation coefficient, large shareholder would increase risk level of this company with investment increase. v.v.This paper has two main innovation places:First, this paper researches large shareholder activism with investment shareholder research.Second, when two companies, belong to the investment portfolio of large shareholder, have positive correlation. The relationship of risk and investment proportion of one company changed with this company's risk level.Another contribute of this paper is provide an activism model of people who have investment portfolio with high asset special character. This model is not only used to study large shareholder activism, just also is used to study manager.In a word, this paper provides a new study model to large shareholder activism. But study has some limitations. The mathematics consequence method is simple, the number of swatches of study isn't enough. There is much limit of study object in multivariate linear regression analysis. These problems need more improve in the future.