中文摘要：

【目的】建立一种基于结构风险最小、既反映样本集动态特征又体现环境因子影响的高精度非线性多维时间序列预测方法。【方法】耦合支持向量机回归（SVR）和带受控项的自回归模型（CAR），以留一法基于MSE最小原则实施模型定阶和变量筛选，以一步预测法检验新模型SVR-CAR的有效性，并通过强制汰选给出各保留变量对预测的相对重要性次序。【结果】3个农业科学实例验证表明，SVR-CAR在7种参比模型中预测精度最高，且可更精细地反映样本集的非线性动态特征，依各保留变量对预测的相对重要性次序及其动态变化可赋予保留变量部分解释能力。【结论】SVR-CAR是一种基于SVR并融合时间序列分析和回归分析的非线性多维时间序列分析方法，具结构风险最小、非线性、适于小样本，能有效克服过拟合、维数灾和局极小，非线性定阶和非线性筛选变量，自动选择核函数及其相应参数，泛化推广能力优异、预测精度高等诸多优点，在农业科学、生态学、经济学等领域有广泛应用前景。

英文摘要：

[Objective] To construct a novel nonlinear multidimensional time series approach based on structural risk minimization, which shows the dynamic characteristics of sample set as well as the effect of environmental factors. [Method] Integrated controlled autoregressive （CAR） into support vector machine regression （SVR）, a novel nonlinear multidimensional time series model named as SVR-CAR was proposed. After estimated the order and screened the variable by leave-one-out method based on the minimum mean square error, the reliability of SVR-CAR was validated by one-step prediction method. [Result] The prediction results of the agricultural sample set showed that SVR-CAR had the highest prediction precision in all reference models, characterized the nonlinear dynamic of the sample sets subtly, and explained the effect of variable to prediction partly according to the order of the restrained variable screened compulsorily. [Conclusion] As a novel nonlinear multidimensional time series analysis approach integrated CAR into SVR, SVR-CAR had the advantages of structural risk minimization, non-linear characteristics, avoiding the over-fit, strong generalization ability and high prediction precision, etc. SVR-CAR, can be widely used in the prediction area of agriculture, ecology and economics.