SCGM11C,预测精度,灰色模型," /> SCGM11C,预测精度,灰色模型,"/> 基于等维动态马尔科夫SCGM(1,1)<sub>C</sub>模型的黄金价格预测
img

QQ群聊

img

官方微信

高级检索

黄金科学技术 ›› 2020, Vol. 28 ›› Issue (1): 158-166.doi: 10.11872/j.issn.1005-2518.2020.01.095

• 采选技术与矿山管理 • 上一篇    

基于等维动态马尔科夫SCGM(1,1)C模型的黄金价格预测

王梅(),陈建宏(),杨珊   

  1. 中南大学资源与安全工程学院,湖南 长沙 410083
  • 收稿日期:2019-06-26 修回日期:2019-08-08 出版日期:2020-02-29 发布日期:2020-02-26
  • 通讯作者: 陈建宏 E-mail:1816057034@qq.com;cjh@263.net
  • 作者简介:王梅(1994-),女,陕西延安人,硕士研究生,从事矿业经济研究工作。1816057034@qq.com
  • 基金资助:
    国家自然科学基金青年基金项目“基于人工智能的矿山技术经济指标动态优化研究”(51404305)

Gold Price Forecast Based on the Equal Dimensional Dynamic Markov SCGM(1,1)C Model

Mei WANG(),Jianhong CHEN(),Shan YANG   

  1. School of Resources and Safety Engineering,Central South University,Changsha 410083,Hunan,China
  • Received:2019-06-26 Revised:2019-08-08 Online:2020-02-29 Published:2020-02-26
  • Contact: Jianhong CHEN E-mail:1816057034@qq.com;cjh@263.net

摘要:

为了提高黄金价格预测精度,提出等维动态马尔可夫SCGM11C预测模型,引入取新去旧的数据处理方法,使用等维动态实现数据优化。等维动态马尔可夫SCGM11C预测模型是将等维动态SCGM11C模型与马尔可夫链结合起来,在等维动态SCGM11C模型的预测结果上再进行状态划分与转移,重新得到预测值。选取2018年1月~2019年4月共16组黄金价格数据,将动态等维的维数定为13,数据处理时选用2018年1月~2019年1月的13个黄金价格数据,预测2019年2月的黄金价格,再依次预测2019年3月和4月的黄金价格。以2019年2~4月的3个黄金价格预测数据作为拟合,预测2019年5月的黄金价格。通过比较灰色SCGM11C预测模型、等维动态SCGM11C模型与等维动态马尔可夫SCGM11C预测模型的精度,可知等维动态SCGM11C预测模型的精度较SCGM11C模型有所提高,等维动态马尔可夫SCGM11C模型的拟合精度最高,达到一级,相对误差平均值为0.85%,符合预测要求,应用该模型预测的2019年5月的黄金价格为1 314.78美元/盎司,实际黄金价格为1 295.55美元/盎司,价格较为接近。

关键词: 黄金价格, 等维动态, 马尔可夫, SCGM11C')">SCGM11C, 预测精度, 灰色模型

Abstract:

In order to improve the accuracy of gold price prediction,an equal dimensional dynamic Markov SCGM(1,1)C forecasting model was proposed.Prediction has high requirements for the selection of data,and the latest data can improve the prediction accuracy.The equal dimensional dynamic Markov SCGM(1,1)C model is a composite model which combines the equal dimensional dynamic SCGM(1,1)C model with the Markov chain.On the basis of the prediction results of the equal dimensional dynamic SCGM(1,1)C,the grey fitting accuracy index is divided into states,and the state of the monthly gold price is determined.On this basis,the next transition direction is determined according to the transition probability matrix,and finally the predicted data is obtained.In this paper,the data processing method of take the new one and remove the old one was introduced,and the equal dimension dynamic data optimization was used.Because the grey SCGM(1,1)Cprediction model is also a grey model,the grey model is characterized by less original data,so a large number of original values are not needed in this paper.A total of 16 groups of gold price data from January 2018 to April 2019 were selected,and the dimension of dynamic equal dimension was determined to be 13.When SCGM(1,1)Cmodel data were processed,13 gold price data from January 2018 to January 2019 were selected to predict the gold price in February 2019,and then the gold price of March 2019 and April 2019 was predicted as above.The prediction data from February 2019 to April 2019 were used as fitting data to observe whether the accuracy of the prediction model is the best.The grey SCGM(1,1)Cmodel was predicted directly with all 16 known data.By comparing the grey SCGM(1,1)Cprediction model,the equal dimensional dynamic SCGM(1,1)Cmodel and the equal dimensional dynamic Markov SCGM(1,1)Cprediction model it is know that the accuracy of the equal dimensional dynamic SCGM(1,1)Cmodel is higher than the SCGM(1,1)Cmodel.The fitting accuracy of the equal dimensional dynamic Markov SCGM(1,1)Cis the highest,reaching the first order,the average relative error is 0.85%,which meets the prediction requirements,and the gold price in May 2019 is predicted to be $1 314.78.Although the grey SCGM(1,1)Cmodel has the lowest accuracy,it is simple to calculate and all the predicted values can be obtained by one calculation.The equal dimensional dynamic Markov SCGM(1,1)Cmodel is the most complex,but its predict results are the most accurate.Compared with the neural network and other methods,the equal dimensional dynamic Markov SCGM(1,1)Cmodel is simpler,so the model can be used to predict the gold price.The gold price in May 2019 is $1 295.55.Which Contrast with the predict is very close.

Key words: gold price, equal dimensional dynamic, Markov, SCGM(1,1)C, forecast precision, gray model

中图分类号: 

  • TD679

表1

等维动态预测算法过程"

步骤输入值预测值
1x(0)(1),x(0)(2),?,x(0)(13)x(0)(14)
2x(0)(2),x(0)(3),?,x(0)(14)x(0)(15)
???
n+1x(0)(n+1),x(0)(n+2),?,x(0)(n+13)x(0)(n+14)

表2

残差精度划分"

相对误差平均值范围/%精度等级相对误差平均值范围/%精度等级
0~1一级5~10三级
1~5二级10~20四级

表3

2018年1月~2019年4月的黄金价格"

时间价格/(美元·盎司-1时间价格/(美元·盎司-1
2018-011 345.12018-091 187.2
2018-021 317.82018-101 215.0
2018-031 323.82018-111 217.6
2018-041 313.22018-121 279.0
2018-051 305.32019-011 323.2
2018-061 250.42019-021 319.2
2018-071 220.92019-031 295.4
2018-081 202.42019-041 282.3

图1

2018年1月~2019年4月的黄金价格"

表4

灰色SCGM(1,1)C模型黄金价格预测结果"

时间黄金价格/(美元·盎司-1原始值/预测值相对误差/%
原始值预测值
2018-011 345.11 281.31.0497-4.7
2018-021 317.81 277.91.0312-3.0
2018-031 323.81 274.51.0386-3.7
2018-041 313.21 271.11.0330-3.2
2018-051 305.31 267.71.0296-2.9
2018-061 250.41 264.40.98891.1
2018-071 220.91 261.00.96813.3
2018-081 202.41 257.60.95614.6
2018-091 187.21 254.30.94655.7
2018-101 215.01 251.00.97123.0
2018-111 217.61 247.60.97592.5
2018-121 279.01 244.31.0278-2.7
2019-011 323.21 241.01.0662-6.2
2019-021 319.21 237.31.0661-6.1
2019-031 295.41 234.41.0494-4.7
2019-041 282.31 231.11.0415-3.9
2019-05-1 250.3825--

表5

等维动态SCGM(1,1)C模型黄金价格预测结果"

时间黄金价格/(美元·盎司-1相对误差/%
实际值预测值
平均相对误差/%-3.66
2019-021 319.21 219.05-7.5
2019-031 295.41 270.34-1.5
2019-041 282.31 250.55-2.5

表6

2018年1月至2019年1月等维动态SCGM(1,1)C模型黄金价格预测结果统计"

时间黄金价格/(美元·盎司-1原始值/预测值相对误差/%
原始值预测值
2018-011 345.11 266.61.0620-5.8
2018-021 317.81 263.41.0430-4.1
2018-031 323.81 260.31.0504-4.7
2018-041 313.21 257.11.0446-4.2
2018-051 305.31 253.91.0410-3.9
2018-061 250.41 250.81.00000.03
2018-071 220.91 247.70.97852.1
2018-081 202.41 244.50.96623.5
2018-091 187.21 241.40.95634.5
2018-101 215.01 238.30.98121.9
2018-111 217.61 235.20.98581.4
2018-121 279.01 232.11.0380-3.6
2019-011 323.21 229.01.0766-7.1

表7

依据灰拟合精度指标的状态划分"

编号灰拟合精度指标状态划分编号灰拟合精度指标状态划分
E10.9500~0.9700E41.0000~1.0420
E20.9700~0.9850E51.0420~1.1200
E30.9850~1.0000

表8

等维动态马尔可夫SCGM(1,1)C模型黄金价格预测结果"

时间黄金价格/(美元·盎司-1相对误差/%模型精度等级
实际值预测值
平均相对误差/%0.85一级
2019-021 319.21 317.73-0.11一级
2019-031 295.41 315.721.50二级
2019-041 282.31 294.300.94一级

图2

黄金价格预测值与原始值拟合对比图"

1 刘亚非,陈燕武.试析黄金市场的灰色—马尔可夫预测[J].企业导报,2011(20):224-226.
Liu Yafei,Chen Yanwu.An analysis of grey Markov forecasting in gold market[J].Guide to Business,2011(20):224-226.
2 林石洁,张卫萍.黄金价格影响因素分析[J].中国有色金属,2018(17):64-65.
Lin Shijie,Zhang Weiping.Analysis of the factors influencing gold price[J].China Nonferrous Metals,2018(17):64-65.
3 张延利,张德生.基于动态数据驱动的改进灰色马尔科夫模型黄金价格预测[J].数学的实践与认识,2016,46(13):31-38.
Zhang Yanli,Zhang Desheng.Improves grey Markov model forecasting the price of gold based on dynamic data driven[J].Mathematics in Practice and Theory,2016,46(13):31-38.
4 陈晓珊,田良辉,韩晓茹.黄金价格预测分析与研究[J].佛山科学技术学院学报(自然科学版),2018,36(4):6-10.
Chen Xiaoshan,Tian Lianghui,Han Xiaoru.Forecasting and analyzing gold price[J].Journal of Foshan University(Natural Science Edition),2018,36(4):6-10.
5 吕海侠.基于相关系数变权重组合模型的黄金价格预测[J].黄金,2017,38(7):3-5.
Haixia Lü.Gold price prediction based on correlation coefficient and variable weight combination model[J].Gold,2017,38(7):3-5.
6 张均东,刘澄,孙彬.基于人工神经网络算法的黄金价格预测问题研究[J].经济问题,2010(1):110-114.
Zhang Jundong,Liu Cheng,Sun Bin.The study on the application of ANFIS in stock index prediction[J].On Economic Problems,2010(1):110-114.
7 张坤,郁湧,李彤.小波神经网络在黄金价格预测中的应用[J].计算机工程与应用,2010,46(27):224-226,241.
Zhang Kun,Yu Yong,Li Tong.Application of wavelet neural network in prediction of gold price[J].Computer Engineering and Applications,2010,46(27):224-226,241.
8 张品一,罗春燕,梁锶.基于GA-BP神经网络模型的黄金价格仿真预测[J].统计与决策,2018,34(17):158-161.
Zhang Pinyi,Luo Chunyan,Liang Si.Gold price simulation prediction based on GA-BP neural network model[J].Statistics and Decision,2018,34(17):158-161.
9 景志刚,施国良.基于小波分析的LS-SVM—ARIMA组合模型的黄金价格预测[J].黄金,2017,38(5):5-8,14.
Jing Zhigang,Shi Guoliang.Gold price prediction using combined LS-SVM and ARIMA model based on wavelet analysis[J].Gold,2017,38(5):5-8,14.
10 许贵阳.基于灰色预测方法的中国黄金期货价格预测模型[J].黄金,2014,35(1):8-11.
Xu Guiyang.Forecasting model of China’s gold futures price based on gray prediction method[J].Gold,2014,35(1):8-11.
11 刘成军,杨鹏,吕文生,等.灰色—马尔科夫复合模型在黄金价格预测中的应用[J].有色金属(矿山部分),2013,65(1):7-11.
Liu Chengjun,Yang Peng,Wensheng Lü,et al.Application of Grey-Markov composite model in forecasting gold price[J].Non-ferrous Metals(Mining Section),2013,65(1):7-11
12 张延利,杨丽.基于改进GM(1,1)模型的黄金价格预测[J].黄金,2015,36(7):6-8.
Zhang Yanli,Yang Li.Gold price prediction based on improved GM(1,1) model[J].Gold,2015,36(7):6-8.
13 Baur D G,Beckmann J,Czudaj R.A melting pot — Gold price forecasts under model and parameter uncertainty[J].International Review of Financial Analysis,2016,48:282-291.
14 Pierdzioch C,Rülke J,Stadtmann G.A note on forecasting the prices of gold and silver:Asymmetric loss and forecast rationality[J].The Quarterly Review of Economics and Finance,2013,53(3):294-301.
15 兰建义,乔美英,周英.煤矿事故预测的马尔可夫SCGM(1,1)C模型的建立与应用[J].安全与环境学报,2016,16(5):6-9.
Lan Jianyi,Qiao Meiying,Zhou Ying.Establishment and application of Markov SCGM(1,1)C model for accidents forecast in coal mining practice[J].Journal of Safety and Environment,2016,16(5):6-9.
16 刘思峰,谢乃明.灰色系统理论及其应用[M].北京:科学出版社,2008.
Liu Sifeng,Xie Naiming.Grey System Theory and Its Application[M].Beijing:Science Press,2008.
17 杨珊,明俊桦,周智勇.基于改进的非线性GM(1,1)模型的职业病预测研究[J].中国安全生产科学技术,2018,14(1):111-116.
Yang Shan,Ming Junhua,Zhou Zhiyong.Study on prediction of occupational diseases based on improved nonlinear GM(1,1) model[J].Journal of Safety Science and Technology,2018,14(1):111-116.
18 姜翔程,陈森发.加权马尔可夫SCGM(1,1)C模型在农作物干旱受灾面积预测中的应用[J].系统工程理论与实践,2009,29(9):179-185.
Jiang Xiangcheng,Chen Senfa.Application of weighted Markov SCGM(1,1)C model to predict drought crop area[J].Systems Engineering—Theory & Practice,2009,29(9):179-185.
19 唐俊勇,田鹏辉,王辉.基于马尔可夫链与服务质量的网络可用性[J].计算机应用,2018,38(12):3518-3523,3528.
Tang Junyong,Tian Penghui,Wang Hui.Network availability based on Markov chain and quality of service[J].Journal of Computer Applications,2018,38(12):3518-3523,3528.
20 冉延平,何万生,雷旭晖,等.应用灰色GM(1.1)模型及其改进模型预测渭河天水段水质[J].水资源与水工程学报,2011,22(5):88-91.
Ran Yanping,He Wansheng,Lei Xuhui,et al.Application of GM(1,1) model and improved model to predict the water quality of Weihe River in Tianshui section[J].Journal of Water Resources and Water Engineering,2011,22(5):88-91.
21 徐建新,杨杰.煤矿百万吨死亡率动态无偏灰色马尔科夫预测[J].中国安全科学学报,2012,22(3):122-127.
Xu Jianxin,Yang Jie.Application of dynamic unbiased grey Markov model in prediction of death rate per million-ton coal[J].China Safety Science Journal,2012,22(3):122-127.
22 李大伟,徐浩军,刘东亮,等.改进的灰色马尔柯夫模型在飞行事故率预测中的应用[J].中国安全科学学报,2009,19(9):53-58.
Li Dawei,Xu Haojun,Liu Dongliang,et al.Improved grey Markov model and its application in prediction of flight accident rate[J].China Safety Science Journal,2009,19(9):53-58.
23 杨灿生,黄国忠,陈艾吉,等.基于灰色—马尔科夫链理论的建筑施工事故预测研究[J].中国安全科学学报,2011,21(10):102-106.
Yang Cansheng,Huang Guozhong,Chen Aiji,et al.Research on construction accident forecast based on Gray-Markov theory[J].China Safety Science Journal,2011,21(10):102-106.
[1] 韩旭,杨珊,王喜梅. 基于商空间的黄金价格SVM模型预测[J]. 黄金科学技术, 2020, 28(1): 148-157.
[2] 李夕兵,朱玮,刘伟军,张德明. 基于主成分分析法与RBF神经网络的岩体可爆性研究[J]. 黄金科学技术, 2015, 23(6): 58-63.
[3] 霍成胜,王成栋,孟军海,白国龙. 基于灰色时序组合模型的基坑监测预测[J]. 黄金科学技术, 2014, 22(5): 79-83.
[4] 张树良. 近期全球黄金市场发展态势[J]. J4, 2009, 17(4): 65-69.
[5] 王忠军, 王永成, 姜维明, 胡世利, 刘金鹏. 降低金矿石回采边界品位分析与应用[J]. J4, 2007, 15(4): 54-57.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
[1] 陈芳芳,张亦飞,薛光. 黄金冶炼污染治理与废物资源化利用[J]. J4, 2011, 19(2): 67 -73 .
[2] 王吉青,王苹,赵晓娟,林乡伟. 黄金生产尾矿综合利用的研究与应用[J]. J4, 2010, 18(5): 87 -89 .
[3] 陈建宏,覃曹原,邓东升 . 基于AHP和物元TOPSIS法的层状岩体巷道冒顶风险评价[J]. 黄金科学技术, 2017, 25(1): 55 -60 .
[4] 彭剑平,沈述保. 绿色矿山建设长效机制与典型案例[J]. 黄金科学技术, 2016, 24(4): 133 -136 .
[5] 宋英昕,宋明春,丁正江,魏绪峰,徐韶辉,李杰,谭现峰,李世勇,张照录, 焦秀美,胡弘,曹佳. 胶东金矿集区深部找矿重要进展及成矿特征[J]. 黄金科学技术, 2017, 25(3): 4 -18 .
[6] 高帅,曾庆栋,于昌明,邢宝山,荆林海,叶杰,范宏瑞,杨奎锋. 遥感及综合物探方法用于山东招远南部隐伏成矿侵入体的空间定位[J]. 黄金科学技术, 2017, 25(5): 1 -10 .
[7] 马凤山,李克蓬,杜云龙,侯成录,李威,张国栋 . 三山岛金矿海底开采F1断裂防突结构可能的破坏形式分析 [J]. 黄金科学技术, 2017, 25(5): 47 -56 .
[8] 宋国政,闫春明,曹佳,郭志峰,鲍中义,刘国栋,李山,范家盟,刘彩杰. 胶东焦家成矿带超千米深部金矿勘查突破及意义——以纱岭矿区为例[J]. 黄金科学技术, 2017, 25(3): 19 -27 .
[9] 陈桥,杨洪英,陈贵民,佟琳琳,牛会群. 尼尔森重选在我国石英脉型金矿选矿工艺中的应用[J]. 黄金科学技术, 2017, 25(5): 73 -79 .
[10] 马凤山,郭捷,李克蓬,卢蓉,张洪训,李威. 三山岛海底金矿开采充填体与顶板岩层的变形监测研究[J]. 黄金科学技术, 2016, 24(4): 66 -72 .