我国主要蔬菜价格波动的关联性分析

选取2004—2017年萝卜、西红柿、黄瓜等12种蔬菜的批发市场周度价格数据,利用广义预测误差方差分解和网络拓扑模型图分析了各品种的关联度。研究结果显示:1)在论证蔬菜价格波动内生性的基础上,发现各蔬菜之间呈现出较高关联度,且不同品种两两之间成对关联度差异明显,其中萝卜存在较强风险脆弱性,黄瓜、茄子等蔬菜则具有较强传染性;2)蔬菜间的相互影响构成蔬菜系统的网络拓扑模型,其中黄瓜、茄子、油菜最具系统重要性,而冬瓜、萝卜的系统重要性最弱。3)调整滞后阶数和预测步数,蔬菜关联度保持在60%～80%之间,依然处于较高水平,结果较为稳健。鉴于此,蔬菜价格波动的防控要把握系统内品种间的关联度,同时注意重点品种的传染性与脆弱性。

Co-integration analysis of main vegetable price fluctuations in China

Based on the weekly wholesale price of 12 vegetables from 2004 to 2017, such as radish, tomato, cucumber, etc., this paper adopted a generalized forecast error variance decomposition and complex network topology model to examine the co-integration of vegetable prices. Results show that: 1) on the basis of the endogenous test, a high order of co-integration among vegetable prices was found through the generalized forecast error variance decomposition model. At the same time, there was a significant difference in the pairwise directional co-integration between different vegetable varieties. Among them, radish price was more vulnerable, and cucumber and eggplant prices were more infectious; 2) the network topology analysis indicate that interactions among varieties in the entire vegetable system formed a complex network pattern: cucumber, eggplant, and rape prices were the most systematically important, while winter melon and radish prices showed a low level of significance; and 3) after adjusting the lag order and the number of prediction steps, the correlation degree was maintained between 60% and 80%, a relatively high degree with stationarity. Therefore, to monitor and control vegetable price fluctuations, this paper suggests to grasp the co-integration among different varieties within the system and to pay more attention to the infectivity and vulnerability of key varieties.