中小企业股票市场化发行定价的半参数自组织模型与实证

随着深圳交易所中小企业板块的正式启动，我国中小企业新股发行的市场化定价方式必将启用。本文采用基于自组织数据挖掘GMDH算法的半参数模型的定价方法，将GMDH算法与半参数模型方法结合起来。不仅利用计算机的自动拟合优选最优复杂度模型，从而大大简化了非参数部分估计。创造了半参数模型崭新的实现方式，并达到了理想的定价效果，而且可以发挥半参数模型的特有优势。进行模型结构分析。经过实证和检验证明了这种模型用于我国股票发行定价的有效性和合理性。为完全市场化股票发行方式下确定股票的发行价格提供了新的方法。

Semiparametric Models Embedded GMDH For Pricing IPOs of Small and Medium-sized Enterprises

With the listing of SME stocks in the Shenzhen stock exchange, pricing Initial Public Offerings (IPOs) for SMEs is an important aspect in finance in china. In this paper, the GMDH algorithm of Self-organizing data mining is developed to estimate semi-parametric models for pricing IPOs. This paper combines the GMDH algorithm of self-organizing data mining and semi-parametric methods. As a result, it not only creates a new realizable way of semi-parameter models and simplifies the estimation of the nonparametric part, but also undertakes the structure analysis. The robustness of the results is tested also.