摘要
对灰色GM(1,N)模型加以改进,提出残差修正GM(1,N)模型。运用残差修正GM(1,N)模型对武汉市住户存款余额进行预测,取得了较好的效果:平均预测误差为11.780 15%,比传统灰色GM(1,N)模型的平均预测误差18.776 38%减小了37.260 80%,近期预测误差更小,2017年预测误差仅为0.604 52%。研究还发现,影响武汉市住户存款余额的因素由大到小依次为城镇化率、城镇居民收入、农村居民收入、人口数量。
The basic principle and improvement measures of grey GM(1, N) model are introduced, and A residual correction GM(1, N) model is proposed. The residual correction GM(1, N) model is used to predict the amount of household deposit balance in Wuhan,and good results are achieved. The average prediction error is 11.780 15%, which is 37.260 80% less than the average prediction error of the conventional grey GM(1, N) model of 18.776 38%. The prediction error is smaller in the near future, and the prediction error is only0.604 52% in 2017. The study also found that the factors influencing the household deposit balance in Wuhn were urbanization rate,urban residents' income rural residents' income and population number from the largest to the smallest.
引文
[1]李明明,王旭皓,孙舒曼.基于ARIMA与季节指数组合模型的居民储蓄存款预测[J].经济研究导刊,2015(2):192-193.
[2]姜洪敏,杨德平.基于灰色马氏链的企业储蓄存款额预测[J].青岛大学学报:自然科学版,2012,19(8):97-101.
[3]祖培福,姬春秋,王秀英.基于GM(1,1)黑龙江居民储蓄存款预测的数学模型[J].数学的实践与认识,2013,43(5):23-27.
[4]刘婧,于洪.山东省城乡居民储蓄存款余额ARIMA预测模型[J].广东外语外贸大学学报,2010,10(7):38.
[5]舒服华.运用改进的灰色GM(1,N)模型预测削匀机磨刀砂轮寿命[J].中国皮革,2016,45(11):11-15.
[6]石萍,唐俊.灰色GM(l,N)模型在经济预测中的应用[J].内蒙古大学学报:自然科学版,2015,46(1):43-47.
[7]慕亚茹,母丽华.基于GM(1,N)改进模型的瓦斯水合物相平衡预测[J].黑龙江科技大学报,2015,25(1):70-74.
[8]樊爱宛,潘中强,王巍.灰色GM(1,N)模型在河南省煤炭需求预测中的应用[J].煤炭技术,2011,30(10):7-9.