基于PCA⁃SSA⁃ELM模型的河北省鸡蛋价格预测及预警  

Prediction and early warning of egg prices in Hebei Province based on PCA⁃SSA⁃ELM model

作　　者：李博薇 邸梦醉 魏子鲲 李丽华[2,4,5] 乔宏 赵浩然[2] LI Bowei;DI Mengzui;WEI Zikun;LI Lihua;QIAO Hong;ZHAO Haoran(College of Economics and Management,Hebei Agricultural University,Baoding 071000,China;College of Mechanical and Electrical Engineering,Hebei Agricultural University,Baoding 071000,China;School of Economics,Nankai University,Tianjin 300071,China;Key Laboratory of Broiler Breeding Facility Engineering,Ministry of Agriculture and Rural Affairs,Baoding 071000,China;Hebei Provincial Key Laboratory of Livestock and Poultry Breeding Intelligent Equipment and New Energy Utilization,Baoding 071000,China)

机构地区：[1]河北农业大学经济管理学院,河北保定071000 [2]河北农业大学机电工程学院,河北保定071000 [3]南开大学经济学院,天津300071 [4]农业农村部肉蛋鸡养殖设施工程重点实验室,河北保定071000 [5]河北省畜禽养殖智能装备与新能源利用重点实验室,河北保定071000

出　　处：《黑龙江畜牧兽医》2023年第8期1-7,共7页Heilongjiang Animal Science And veterinary Medicine

基　　金：国家自然科学基金项目(31902209);河北省二期现代农业产业技术体系创新团队建设项目(HBCT2018150208);河北省社会科学基金项目(HB20GL039)。

摘　　要：为了促进河北省蛋鸡产业健康有序发展,研究以2011年1月份—2020年12月份的120组河北省鸡蛋月度价格数据作为训练样本,分别从供给、需求和消费者收入水平的角度选取影响鸡蛋价格的变量,采用主成分分析(PCA)法对样本数据进行降维处理,通过麻雀搜索算法(SSA)对极限学习机(ELM)进行优化,构建了PCA⁃SSA⁃ELM模型,对河北省2021年1月份—2022年5月份共17个月的鸡蛋价格进行预测及预警。结果表明:构建的PCA⁃SSA⁃ELM模型整体拟合效果较好,均方根误差为0.23;2021年1月份—2022年5月份河北省鸡蛋预测价格整体呈上涨趋势,波动范围不大,价格整体稳定,其中出现1次正向重警,1次负向重警,1次正向轻警,1次负向轻警,其余13个月均为无警状态;与真实价格对比,共有14个月的警度预测正确,准确率为82.35%。说明该模型可以较为准确地预测河北省鸡蛋价格。最后笔者对河北省蛋鸡产业提出以下对策建议:健全信息服务平台,做好对鸡蛋产业的服务工作;深挖综合数据,优化鸡蛋价格风险预警指标体系;完善对河北省蛋鸡产业的扶持政策。In order to promote the healthy and orderly development of layer industry in Hebei Province,in this study,120 sets of egg monthly price data of Hebei Province from January 2011 to December 2020 were taken as training samples.Variables affecting egg price were selected from the perspectives of supply,demand and consumer income.Principal component analysis(PCA)was used to reduce the dimension of sample data,and sparrow search algorithm(SSA)was used to optimize the extreme learning machine(ELM).The PCA⁃SSA⁃ELM model was constructed to forecast and warn the egg price in Hebei Province for 17 months from January 2021 to May 2022.The results showed that the overall fit effects of the PCR⁃SSA⁃ELM model constructed was good,and the root⁃mean⁃square error was 0.23,which could predict the egg price in Hebei Province more accurately.The predicted egg prices in Hebei Province from January 2021 to May 2022 showed an overall rising trend,with a small fluctuation range and overall stable price,among which there was one positive heavy alarm,one negative heavy alarm,one positive light alarm,one negative light alarm,the remaining 13 months showed no alarm state.Compared with the real prices,a total of 14 months of alarm prediction was correct;the accuracy of calculation was 82.35%.The results showed that this model can predict egg prices accurately in Hebei Province.Finally,the author put forward the following countermeasures and suggestions to the layer industry of Hebei Province:improve the information service platform,and do a good job in serving the egg industry;dig the comprehensive data deeply,optimize the egg price risk warning index system;perfect the egg industry support policy of Hebei Province.

关 键 词：价格预测 价格预警 鸡蛋 麻雀搜索算法 极限学习机 

分 类 号：F323.7[经济管理—产业经济]