基于自回归神经网络的钼铁价格预测分析

Price Forecasting Analysis of Ferric Molybdenum Based on Autoregressive Neural Network

钼铁在工具钢生产企业的原料采购成本中占比20%,其价格的大幅波动给相关企业的生产和经营的稳定性造成了巨大影响。研究钼铁价格波动的规律,对于指导钼铁采购企业制定合适的采购方案、降低采购成本、保证生产经营的稳定性具有举足轻重的作用。通过对钼铁市场信息及钼铁价格影响因子进行分析,根据钼铁价格波动的特点,引入自回归神经网络,建立了适合下游企业钼铁月度价格预测模型和日价格预测模型,证明了自回归神经网络在钼铁价格预测方面的适用性,拓展了钼铁价格预测的思路。

Ferric molybdenum,as a necessary raw material for tool steel production enterprises,accounts for20% of the raw material purchasing cost of tool steel production enterprises.The price of ferric molybdenum fluctuates greatly in recent years,which has caused a great impact on the production and operation stability of related enterprises.The study of ferric molybdenum price fluctuation law plays a pivotal role in guiding ferric molybdenum purchasing enterprises to formulate appropriate procurement plans,reduce procurement costs,and ensure the stability of production and operation.In this paper,the market information of ferric molybdenum and the influencing factors of ferric molybdenum price are analyzed.According to the characteristics of ferric molybdenum price fluctuation,the autoregressive neural network is introduced to establish the monthly price forecast model and the daily price forecast model suitable for downstream enterprises.The applicability of autoregressive neural network in ferro molybdenum price prediction is proved,and the idea of ferro molybdenum price prediction is expanded.The accurate prediction of ferro molybdenum price can provide better decision-making basis for the purchasing,production,operation and other behaviors of related enterprises,which has strong practical and theoretical significance.