Abstract: Because the product of venture capital project has enormous uncertainty, a single venture capital which is restrained by his resource hardly can make accurate evaluation. As a result, the leading venture capital needs to associate with other venture capital to make syndicating investment. By introducing the venture capitalist's costs of risk aversion, costs of effort and strategic benefits, three-stage investment decision model is established in this paper to discuss the problem of syndicating investment decision in venture capital with learning effect of signaling theory, which proceeds from brand-new perspective of information economics. The leading VC use Bayesian rule to amend the prior belief based on the signal of following VC's project evaluation, and then adjusts his own project evaluation. Through the learning effect of signaling between the partner syndicating investments to seek the cooperation interval of syndication in venture capital, the success rate of venture capital can be increased then, The conclusion has important reference value to the practice of syndicate investment decision-making. Study variables and parameters involved in the decision-making process to the empirical data in this article are obtained mainly from the two ways, firstly, using likert five-point scale to measure the basic variables, which are the venture capitalist's costs of risk aversion and effort, strategic benefits. Other experience variables such as project success probability and output level type, need to be estimated based on the venture capitalists ability and industry experience. In the numerical example, through the analysis of the syndicate investment decision-making process of the two projects, the conclusion is completely consistent with the theoretical model, verified the theoretical model is scientific and validity.

Key words: venture capital, syndicating investment, signaling, learning effect, Bayesian rule