非线性多元样条回归预报模型

NONLINEAR MU LTIPLE SPLINE REGRESSION MODEL

  • 摘要: 针对异常复杂的非线性预报系统,提出非线性多元样条回归预报模型。该模型既保留了线性统计预报中多元分析,逐步筛选预报因子及显著性检验的理论和方法,又吸取了样条函数分段拟合,按需要裁剪以适应任意曲线连续变化的优点,具有处理复杂非线性预报系统的功能。试验结果表明,该模型具有良好的模拟和预报能力,值得进一步深入探讨和应用。

     

    Abstract: The nonlinear multiple spline regression model was advanced. The model which retains multivariate analysis, screening factories and significance test of linear statistical prediction, and absorbs the advantage of splines that can simulate data stagewisely and can autoadapte any change of curve can deal with the situation which is abnormally complicated. The results indicate that the model can efficiently simulate and predict nonlinear system, and it is worthy to be studied and applied.

     

/

返回文章
返回