基于SARIMA模型的我国水产品消费价格指数预测  

Forecast of Consumer Price Index for Aquatic Products in China Based on SARIMA Model

作　　者：杨振昊 张俊波 杨晨星[5] 陈新军 YANG Zhenhao;ZHANG Junbo;YANG Chenxing;CHEN Xinjun(College of Marine Sciences,Shanghai Ocean University,Shanghai 201306,China;National Oceangoing Fishery Engineering Technology Research Center,Shanghai 201306,China;National Experimental Teaching Demonstration Center of Aquatic Science(Shanghai Ocean University),Shanghai 201306,China;Key Laboratory of The Ministry of Education for sustainable Development of Ocean Fishery Resources,Shanghai 201306,China;College of Economics and Management,Shanghai Ocean University,Shanghai 201306,China)

机构地区：[1]上海海洋大学海洋科学学院,上海201306 [2]国家远洋渔业工程技术研究中心,上海201306 [3]水产科学国家级实验教学示范中心(上海海洋大学),上海201306 [4]大洋渔业资源可持续开发省部共建教育部重点实验室,上海201306 [5]上海海洋大学经济管理学院,上海201306

出　　处：《海洋湖沼通报》2021年第2期131-138,共8页Transactions of Oceanology and Limnology

基　　金：上海市青年东方学者项目(QD2017038);远洋鱿钓渔业管理制度(D-8021-19-5003);我国远洋鱿钓渔业产品流通与加工研究(A1-0203-00-300101);全球鱿鱼生产与贸易研究(A2-0203-00-100363)资助。

摘　　要：水产品消费价格变动会引起消费需求量和渔民收入变化的双重反应,准确预测价格指数对水产品市场具有重要的预警作用。本文基于1995年1月到2019年6月我国水产品类居民消费价格指数历史数据,通过Box-Jenkins方法构建季节性差分自回归滑动平均(SARIMA)模型,利用2019年7月至12月水产品消费价格指数进行模型验证,并预测了2020年1月至6月我国水产品价格指数波动。结果显示SARIMA(1,1,2)(1,1,1)^(12)模型预测结果与实际值之间的均方根误差(RMSE)、平均绝对误差(MAE)和平均绝对百分比误差(MAPE)都较小,MAPE值小于10%;2020年上半年我国水产品消费价格指数大致呈稳定状态,在97.0到97.7之间上下波动。研究结果表明本文所构建的SARIMA模型拟合结果较好、预测结果精度较高,可以较好地解释我国水产品消费价格指数变动的季节性与趋势性,为相关管理部门与单位预测水产品消费价格提供科学有效的方法,对合理应对我国水产品市场价格波动具有一定的参考价值。Changes in the consumption price of aquatic products will affect demand and fishermen’s income.Accurate prediction of consumer price index(CPI)provides important information regarding the aquatic product market.Based on the historical data of aquatic products CPI from January 1995 to June 2019,this paper constructs a seasonal autoregressive integrated moving average(SARIMA)model applying the Box-Jenkins method.The aquatic product CPI from July to December 2019 are used to verify the optimal model SARIMA(1,1,2)(1,1,1)^(12);and the model is utilized to predict the CPI for aquatic products in the first half of 2020.Results show that the root mean square error(RMSE),mean absolute error(MAE)and mean absolute percentage error(MAPE)between the predicted and the actual values of aquatic product CPI from July to December 2019 are all small with the MAPE value less than 10%;and the predictive CPI for aquatic products from January to June 2020 keeps relatively stable.This explains that SARIMA(1,1,2)(1,1,1)^(12) established in this paper performs well with high predictive accuracy,which could better clarify the fluctuation of aquatic products price levels and provide a scientific and effective method for predicting the CPI of aquatic products.

关 键 词：SARIMA模型 水产品 价格预测 

分 类 号：F307.4[经济管理—产业经济]