卷烟烟气焦油量的ARMA预测模型研究.pdfVIP

安徽农业科学．Journal of Anhui Agri．Sci．2008，36(14)：5696-5698 责任编辑 张杨林 责任校对 卢瑶 卷烟烟气焦油量的ARMA预测模型研究 陆鸣1，李东亮2许自成2刘秀彩 ，许寒春 (1．福建中烟技术中心，福建厦门361022；2．河南农业大学国家烟草栽培生理生化研究基地，河南郑州450002) 摘要 『目的1为计算机辅助配方研究提供理论依据。『方法]以同一品牌不同批次卷烟的烟气焦油量为研究对象，采用时间序列分析 法，建立了卷烟烟气焦油量的预测模型，并进行了模型预测验证o[~ ．-]ARMA(2，2)模型的AIC,LF和FPE值在各模型中均最小，所 以选择ARMA(2，2)为卷烟烟气焦油量的预测模型，即：(1—1．622 q-l+O．844 q )y( )=(1—1．836q +1．O2 q-Z e(t)。根据对模型残差序列 进行的白噪声检验判定，建立的ARMA(2，2)模型是显著有效模型，模型预测验证表明模型预测精度达9％51％，平均相对误差为 O．49％，属于一级(优等)模型。时间序列一般用于短期预测，不能用于长期预测。『结论]该研究建立的卷烟烟气焦油量的ARMA(2，2) 预测模型的预测精度高、误差小 可以用于卷烟烟气焦油量的短期预测。 关键词 卷烟；烟气焦油量；ARMA模型；时间序列分析 中图分类号 $572 文献标识码 A 文章编号 0517--6611(2008)14—05696—03 ARMA Forecast Model of Tar of Cigarette Smoke LU Ming et al(Technology Center of China Tobacco Fu3iJan Industrial Corporation，Xiamen，VujJan 361022) Abstract 『Objective】The study aimed to provide a base for computer—aided formula design．『Method]With the tar of cigarettes from differenl， batches of cigarettes with the same trademark as tested materials and by using time series sequence method，the forecast model of tar of cigarette smokewas built and validated．『Result]Amongforecastmodels，the values ofAIC，￡，and 啊were allleastinARMA(2，2)model，namely【1- 1．622 q一 +O．844 q Y(t)=(1—1．836 q一 +1．02 q e(t)，llso it was selected as forecast model for tar ofcigarette smoke．The established ARMA (2，2)model was significantly effective model according to autocorrelation check ofthe residuals on the model residual sequence。The forecast Drecision ofARMA(2，2)model reached 99．51％and the average relative error was 0．49％，so the ARMA(2，2)model belonged to the first grade model fexce~ent nodel1．However．time series model was usually used for short--term forecast but not for long germ forecast．[Conclusion]The forecast model of tar of cigarettes smoke built in this study had a better precision．1ower error and could beused for shoa~erm forecast of the tar 0f cigarettes smoke． Key words Cigarettes