基于ARMAV模型的国内海洋捕捞与海水养殖产量的分析

The analysis of marine capture production and mariculture production in China by ARMAV models

为了揭示国内海洋捕捞和海水养殖产量的相关关系,准确地跟踪并预测海洋捕捞和海水养殖产量的短期未来趋势,利用时序分析方法对1954—2006年国内海洋捕捞和海水养殖产量数据建立了多维自回归滑动平均(ARMAV)模型。该方法不仅避免了分别使用自回归滑动平均(ARMA)模型对两序列建模未考虑序列间关系的弊端,还通过数据的先期平稳化处理而使得算法的运用更具有针对性。图像与误差计算结果均表明,用本研究中给出的ARMAV(2,1,2)算法对两序列进行跟踪及预测具有效性。

The auto - regressive moving average vector （ARMAV） models are established by the data of marine capture production and mariculture production in China from 1954 to 2006 to reveal the relationship between the marine capture production and mariculture production in China, tracking and forecasting accurately the short - term trend in future. The use of the ARMAV models here not only prevents the sickness of un - touching the relationship between the marine capture production and the mariculture production when using auto -regressive moving average （ARMA） models for the two sets of data respectively, but also makes the algorithm have pertinence after smoothness disposal of the data in advance. Both of the results of the images and the errors demonstrate the validity of the algorithm when tracking and forecasting the two data set by the ARMAV（2,1,2） models given in the paper.