Abstract This thesis is composed of three models that imply the impact of managers¡¦ characteristics on investment under asymmetric information. The first one regards insider as managers and formulates a model to explain the positive relationship between cash flow and capital expenditure of a firm, and tries to synthesize the ¡§asymmetric information hypothesis¡¨(Myers and Majluf, 1984) and the ¡§free cash flow hypothesis¡¨(Jensen, 1986) by insider ownership. The finding demonstrates that in instances with low percentage of insider ownership, the free cash flow hypothesis will better explain the positive relationship between cash flow and capital expenditure and will have the phenomenon of over-investment. On the other hand, when the percentage of insider ownership is high, the asymmetric information hypothesis is better suited to explain this relationship and will have the phenomenon of the under-investment. The second one formulates a model to synthesize the ¡§reputation effect¡¨ and ¡§asymmetric information hypothesis¡¨ through considering the outsider investors¡¦ evaluation of the firms in terms of firms¡¦ reputation and firms¡¦ private information. This study concludes that the good type firms with low reputation will show the behavior of under-investment and the bad type firms with high reputation will have the phenomenon of over-investment. Moreover, the model demonstrates that both the phenomena of under-investment and of over-investment are caused by the conflict between the firms and the outsider investors. At last, this study implies that the effect of reputation has an influence on the choice of financial tools for the good type firms but does not have an influence on that of the bad ones. This study presents a general model to explain two types of investment inefficiency under the effect of reputation in a reasonable mode. The last one formulates a model to synthesize the ¡§bias effect¡¨ and ¡§reputation effect¡¨ through considering the fact that the CEO in the interest of firm is in favor of a certain project and that junior managers concern their reputation. This study concludes that the CEO¡¦s bias will influence the project that the managers suggest and does not necessarily lead to the direction of bias. The untalented managers will be affected more seriously than talented managers. Moreover, the model combines ¡§bias effect¡¨ with ¡§asymmetric information hypothesis¡¨ and implies that the bias can alleviate the problem of under-investment under certain circumstances. This finding shows that the bias is not always a negative factor of investment efficiency.