中文摘要：

中止决策能力是影响风险投资公司长期绩效的一个重要指标，然而现有中止决策方法忽视了所投资新兴技术企业发展过程中释放信息的影响。针对该问题，本文首先结合新兴技术企业特点，归纳分析了新兴技术企业发展过程中释放的信息；其次以风险投资家观察到的好（坏）信号作为二元学习信号，就好（坏）信号随新兴技术企业发展存在的三种状态，从贝叶斯后验估计的角度提出了信号学习模型；继而依据蕴含了风险投资家风险态度的“心理阀值”确定外生的中止决策点；最后进行了算例分析。本文提出的信号学习模型是基于贝叶斯后验估计的动态模型，反映了信息的动态发展对后续投资决策的影响。该模型为风险投资家及时、准确做出中止决策提供理论参考，也为创业者常见的“窗饰效应”作了合理的解释。

英文摘要：

The ability to make termination decision is an important index that influences on the venture capital firms＇ long term performance. However, present termination decision methods ignore the information release effect during the emerging technology firm development process. Aiming at this problem, the information release during the emerging technology firm development process is induced and analyzed firstly; then the good （bad） signal observed by venture capitalist is chosen as the binary learning signal, three conditions that the good （bad） signal observed along with the emerging technology firm development are considered, from the view of Bayesian posterior estimate, a signal learning model is proposed ; the exogenous optimal stopping point is decided according to venture capitalist＇ s psychological threshold which reflects venture capitalist＇ s altitude to risk farther more; and an example is given at the end. The model is a dynamic one based on the Bayesian posterior estimate; and it reflects the information dynamic development influences on the following decision. The model provides a theoretical reference for venture capitalist＇ s termination decision in time, and it also gives a reasonable explanation for the enterpriser＇s ＂window dressing＂ behavior.