基于SVR和CAR的多维时间序列分析及其在生态学中的应用

基于支持向量回归（SVR）并融合带受控项的自回归模型（CAR），建立了一种既反映样本集动态特征又体现环境因子影响的非线性多维时间序列分析预测方法（SVR-CAR）。用一步预测法对两个生态学样本集的预测结果表明，SVR-CAR在所有参比模型中预测精度最高，并具结构风险最小、非线性、避免过拟合、泛化推广能力优异等诸多优点。SVR-CAR在生态学、农业科学、经济学等多维时间序列预测领域有较广泛的应用前景。

Multidimensional time series analysis based on support vector regression and controlled autoregressive and its application in ecology

Based on support vector regression (SVR) and controlled autoregressive (CAR), we proposed a new non-linear multidimensional time series method named SVR-CAR that can show the dynamic characteristics of sample set as well as the effect of environmental factors. To evaluate the performance of SVR-CAR, we compared its predictions with those of four other commonly-used methods, using two sets of real-world data and one-step prediction. The results showed that SVR-CAR had the highest accuracy in prediction among the five methods, and had the advantages of structural risk minimization, non-linear characteristics, avoiding over-fit, and strong capacity for generalization. SVR-CAR has the potential to be widely used for predictions involving multidimensional time series data in ecology, agricultural sciences and economics.