基于人类主体实验的学习模型校验

The Learning Modelvalidation Based on the Subject Experiment

本文采用人类主体实验的方法获得符合本国投资者学习过程的股价序列,来校验计算实验金融里Roth-Erve学习模型中的遗忘参数和类比参数。通过在一定区间内改变学习模型参数的大小,得到拟合性相对高、适应本国投资者学习过程的模型参数。研究表明:人类主体实验得到收敛和阻尼振动的两类股价序列;用学习模型对两类股价序列进行模拟,找到拟合性高的解,并分析其理论意义;在给定的区间内,通过拟合性高低的比较,可以找到相对本国投资者适应性更高的参数值。这为计算实验金融的应用,提供了更适应本国投资者的学习模型。

This paper uses subject experiment method to get the shares sequence of domestic investors and check the forgotten parameter and analogy parameters in Roth-Erve learning model used in Agent-based Model for Finance.Through changing the size of parameters in a given range,to get the model parameters which are more adoptive to the domestic investors and the fitness is relatively high.Research has shown that subject experiment concludes in convergence and damping vibration of two classes of shares sequence;using the learning model to simulate a sequence of two classes of shares,then we can find solution with high fitness,and analyze its meaning of theory;in a given interval,by comparing the fitness,we can also find the parameter values which are more adaptive for domestic investors.For the application of Agent-based Model,the research provides a more adaptive learning model of domestic investors.