基于神经网络的医药高科技投资风险评价模型

Medicine Venture Capital Investment Evaluation Model Based on ANN

作为风险投资三大领域之一的医药高科技风险投资在各国都受到极大重视,吸引着巨大的资金,其投资金额仅次于IT产业高科技风险投资。但经典的风险管理理论由于基于理想化假设及需要历史数据预测的数学方法对于风险投资有着巨大的局限性。本文在建立医药高科技风险投资评价体系的基础上,提出一种基于BP神经网络的适合医药高科技项目风险投资多指标综合评价方法。该方法通过神经网络的自学习、自组织适应能力和强容错性,能准确地按照专家的评定方法进行工作,训练好的神经网络系统将专家评价思想以连接权的方式赋予神经网络,并最终在实际投资中被运用和验证。

With an analysis of the risk of investing in high-tech in medicine projects, this paper proposes a multi target synthetic evaluation method based on neural network. The method can simulate the expert in evaluating the investing risk, thus avoiding the subjective mistakes in the evaluation process. The simulation test indicates that results achieved by this method are satisfactory.