小样本协整检验的神经网络方法  

Cointegration Test with Small Sample: Comparison of Response Surface and Neural Network Approaches

作　　者：杨宝臣[1] 张世英[1] 

机构地区：[1]天津大学管理学院,天津300072

出　　处：《系统工程理论方法应用》2004年第5期409-413,共5页Systems Engineering Theory·Methodology·Applications

基　　金：国家自然科学基金资助项目(79400014)

摘　　要：采用MonteCarlo模拟方法,给出了小样本容量下协整检验临界值的响应面方程。结果表明,小样本协整检验的临界值不仅与样本容量有关,还依赖于协整检验中的滞后阶数。进一步,以样本容量、变量个数、滞后阶数为输入指标,检验临界值为输出指标建立了神经网络模型。神经网络模型比响应面模型具有更高的精度。In this paper, we first developed the response surface equations for determination of critical value for cointegration test in small sample using Monte Carlo simulation, the results indicate that the critical values for cointegration test are related with the sample size, as well as the lagged order. And then, we established the neural network model for cointegration test with small sample, where the sample size, number of variables, and lagged order were employed as inputs and critical values as outputs. Evidence shows that the neural network approach is more accurate than the response surface model which is commonly used in critical value determinationfor critical values.

关 键 词：协整检验 响应面方程 神经网络 小样本 

分 类 号：O213[理学—概率论与数理统计]