中国管理科学 ›› 2024, Vol. 32 ›› Issue (8): 149-158.doi: 10.16381/j.cnki.issn1003-207x.2022.1937
• • 上一篇
收稿日期:
2022-09-01
修回日期:
2023-02-21
出版日期:
2024-08-25
发布日期:
2024-08-29
通讯作者:
曾波
E-mail:bozeng@ctbu.edu.cn
基金资助:
Shuliang Li,Shuangyi Yang,Bo Zeng(),Wei Meng,Yun Bai
Received:
2022-09-01
Revised:
2023-02-21
Online:
2024-08-25
Published:
2024-08-29
Contact:
Bo Zeng
E-mail:bozeng@ctbu.edu.cn
摘要:
无偏性与自适应性是灰色预测模型的两个重要性质,是研究模型结构及性能的基础。首先,文章以矩阵为工具,从理论上严格证明了两类常见离散灰色预测模型的无偏性。结果表明:双参数离散灰色预测模型DGM(1,1)仅对齐次指数序列具有无偏性,而三参数离散灰色预测模型TDGM(1,1)则对齐次指数/非齐次指数/线性函数序列均具有无偏性。然后,从模型结构角度对TDGM(1,1)模型面向不同特征序列的自适应性进行了分析,验证了模型结构自适应性与无偏性之间的内在联系。最后,应用TDGM(1,1)模型对世界新能源汽车销售量进行建模,并对预测结果进行了比较和分析。本研究对丰富和完善灰色预测理论具有积极意义。
中图分类号:
李树良,杨爽艺,曾波,孟伟,白云. 离散型灰色预测模型的无偏性与自适应性及其应用[J]. 中国管理科学, 2024, 32(8): 149-158.
Shuliang Li,Shuangyi Yang,Bo Zeng,Wei Meng,Yun Bai. Unbiased and Adaptive of Discrete Grey Prediction Model and Its Application[J]. Chinese Journal of Management Science, 2024, 32(8): 149-158.
表1
齐次指数序列的无偏性模拟结果"
TDGM(1,1) | DGM(1,1) | GM(1,1) | NGM(1,1,k) | ||||||
---|---|---|---|---|---|---|---|---|---|
1.08 | — | — | — | — | — | — | — | — | |
1.94 | 1.94 | 0.00 | 1.94 | 0.00 | 1.87 | 3.65 | 1.13 | 41.71 | |
3.50 | 3.50 | 0.00 | 3.50 | 0.00 | 3.32 | 5.21 | 2.21 | 36.79 | |
6.30 | 6.30 | 0.00 | 6.30 | 0.00 | 5.87 | 6.75 | 4.12 | 34.64 | |
11.34 | 11.34 | 0.00 | 11.34 | 0.00 | 10.40 | 8.26 | 7.48 | 34.00 | |
20.41 | 20.41 | 0.00 | 20.41 | 0.00 | 18.42 | 9.75 | 13.43 | 34.21 | |
36.73 | 36.73 | 0.00 | 36.73 | 0.00 | 32.61 | 11.22 | 23.93 | 34.86 | |
66.12 | 66.12 | 0.00 | 66.12 | 0.00 | 57.75 | 12.66 | 42.48 | 35.76 | |
119.02 | 119.02 | 0.00 | 119.02 | 0.00 | 102.26 | 14.08 | 75.24 | 36.78 | |
214.23 | 214.23 | 0.00 | 214.23 | 0.00 | 181.09 | 15.47 | 133.11 | 37.87 | |
0.00 | 0.00 | 9.67 | 36.29 | ||||||
385.61 | 385.61 | 0.00 | 385.61 | 0.00 | 320.67 | 16.84 | 235.33 | 38.97 | |
694.10 | 694.10 | 0.00 | 694.10 | 0.00 | 567.84 | 18.19 | 415.88 | 40.08 | |
1249.38 | 1249.38 | 0.00 | 1249.38 | 0.00 | 1005.50 | 19.52 | 734.82 | 41.19 | |
2248.88 | 2248.88 | 0.00 | 2248.88 | 0.00 | 1780.60 | 20.82 | 1298.20 | 42.28 | |
4047.98 | 4047.98 | 0.00 | 4047.98 | 0.00 | 3153.10 | 22.11 | 2293.30 | 43.35 | |
0.00 | 0.00 | 19.50 | 41.17 | ||||||
0.00 | 0.00 | 13.18 | 38.03 |
表2
非齐次指数序列的无偏性模拟结果"
TDGM(1,1) | DGM(1,1) | GM(1,1) | NGM(1,1,k) | ||||||
---|---|---|---|---|---|---|---|---|---|
4.38 | — | — | — | — | — | — | — | — | |
5.05 | 5.05 | 0.00 | 3.78 | 25.17 | 3.74 | 25.89 | 2.06 | 59.18 | |
5.99 | 5.99 | 0.00 | 4.98 | 16.88 | 4.93 | 17.80 | 3.20 | 46.63 | |
7.31 | 7.31 | 0.00 | 6.56 | 10.21 | 6.48 | 11.34 | 4.66 | 36.24 | |
9.15 | 9.15 | 0.00 | 8.65 | 5.52 | 8.53 | 6.85 | 6.54 | 28.53 | |
11.74 | 11.74 | 0.00 | 11.40 | 2.89 | 11.22 | 4.41 | 8.96 | 23.62 | |
15.35 | 15.35 | 0.00 | 15.02 | 2.17 | 14.76 | 3.85 | 12.08 | 21.31 | |
20.41 | 20.41 | 0.00 | 19.79 | 3.05 | 19.42 | 4.86 | 16.09 | 21.18 | |
27.49 | 27.49 | 0.00 | 26.07 | 5.16 | 25.55 | 7.08 | 21.25 | 22.72 | |
37.41 | 37.41 | 0.00 | 34.36 | 8.16 | 33.61 | 10.16 | 27.89 | 25.46 | |
0.00 | 8.80 | 10.25 | 31.65 | ||||||
51.29 | 51.29 | 0.00 | 45.27 | 11.74 | 13.80 | 10.25 | 36.43 | 28.98 | |
70.73 | 70.73 | 0.00 | 59.66 | 15.66 | 17.76 | 13.80 | 47.42 | 32.96 | |
97.95 | 97.95 | 0.00 | 78.61 | 19.74 | 21.86 | 17.76 | 61.57 | 37.14 | |
136.04 | 136.04 | 0.00 | 103.58 | 23.86 | 25.99 | 21.86 | 79.77 | 41.37 | |
189.38 | 189.38 | 0.00 | 136.49 | 27.93 | 30.05 | 25.99 | 103.19 | 45.51 | |
0.00 | 19.79 | 21.89 | 37.19 | ||||||
0.00 | 12.72 | 14.41 | 33.63 |
表3
线性函数序列的无偏性模拟结果"
TDGM(1,1) | DGM(1,1) | GM(1,1) | NGM(1,1,k) | ||||||
---|---|---|---|---|---|---|---|---|---|
4.80 | — | — | — | — | — | — | — | — | |
8.40 | 8.40 | 0.00 | 11.70 | 39.27 | 11.64 | 38.60 | 5.93 | 8.40 | |
12.00 | 12.00 | 0.00 | 13.63 | 13.59 | 13.57 | 13.06 | 9.83 | 12.00 | |
15.60 | 15.60 | 0.00 | 15.88 | 1.82 | 15.81 | 1.35 | 13.67 | 15.60 | |
19.20 | 19.20 | 0.00 | 18.51 | 3.61 | 18.42 | 4.04 | 17.43 | 19.20 | |
22.80 | 22.80 | 0.00 | 21.56 | 5.42 | 21.47 | 5.84 | 21.13 | 22.80 | |
26.40 | 26.40 | 0.00 | 25.13 | 4.82 | 25.02 | 5.23 | 24.76 | 26.40 | |
30.00 | 30.00 | 0.00 | 29.28 | 2.41 | 29.16 | 2.82 | 28.32 | 30.00 | |
33.60 | 33.60 | 0.00 | 34.11 | 1.53 | 33.98 | 1.12 | 31.83 | 33.60 | |
37.20 | 37.20 | 0.00 | 39.75 | 6.85 | 39.59 | 6.43 | 35.27 | 37.20 | |
0.00 | 8.81 | 8.72 | 10.96 | ||||||
40.80 | 0.00 | 46.32 | 13.52 | 46.14 | 13.08 | 38.65 | 40.80 | 0.00 | |
44.40 | 0.00 | 53.97 | 21.55 | 53.77 | 21.09 | 41.96 | 44.40 | 0.00 | |
48.00 | 0.00 | 62.88 | 31.01 | 62.65 | 30.53 | 45.22 | 48.00 | 0.00 | |
51.60 | 0.00 | 73.27 | 42.00 | 73.01 | 41.50 | 48.43 | 51.60 | 0.00 | |
55.20 | 0.00 | 85.38 | 54.67 | 85.08 | 54.14 | 51.57 | 55.20 | 0.00 | |
0.00 | 32.55 | 32.07 | 5.86 | ||||||
0.00 | 17.29 | 17.06 | 9.14 |
表4
TDGM(1,1)、DGM(1,1)、GM(1,1)及NGM(1,1,k)模型对不同序列的模拟效果比对"
典型数据序列 | 齐次指数序列 | 非齐次指数序列 | 线性函数序列 | ||||||
---|---|---|---|---|---|---|---|---|---|
TDGM(1,1) | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
DGM(1,1) | 0.00 | 0.00 | 0.00 | 8.80 | 19.79 | 12.72 | 8.81 | 32.55 | 17.29 |
GM(1,1) | 9.67 | 19.50 | 13.18 | 10.25 | 21.89 | 14.41 | 8.72 | 32.07 | 17.06 |
NGM(1,1,k) | 36.29 | 41.17 | 38.03 | 31.65 | 37.19 | 33.63 | 10.96 | 5.86 | 9.14 |
表7
世界新能源汽车的模拟值与模拟误差"
年份 | TDGM(1,1) | DGM(1,1) | GM(1,1) | NGM(1,1,k) | |||||
---|---|---|---|---|---|---|---|---|---|
2013 | 21.50 | — | — | — | — | — | — | — | — |
2014 | 33.50 | 34.06 | 1.67% | 58.41 | 74.37% | 57.14 | 70.58% | 32.04 | 4.35% |
2015 | 68.90 | 64.93 | 5.76% | 78.88 | 14.48% | 77.14 | 11.96% | 57.43 | 16.65% |
2016 | 93.40 | 101.17 | 8.32% | 106.51 | 14.03% | 104.13 | 11.49% | 88.31 | 5.45% |
2017 | 139.90 | 143.72 | 2.73% | 143.82 | 2.80% | 140.57 | 0.48% | 125.86 | 10.04% |
2018 | 218.50 | 193.68 | 11.36% | 194.20 | 11.12% | 189.75 | 13.16% | 171.52 | 21.50% |
2019 | 228.40 | 252.32 | 10.47% | 262.24 | 14.81% | 256.15 | 12.15% | 227.06 | 0.59% |
2020 | 328.40 | 321.17 | 2.20% | 354.10 | 7.83% | 345.78 | 5.30% | 294.59 | 10.29% |
6.07% | 19.92% | 17.87% | 9.84% |
1 | Deng J L. The control problem of grey systems[J]. System Control Letter, 1982, 1(5),288-294. |
2 | 刘思峰.灰色系统理论及其应用(第九版)[M].北京:科学出版社,2021. |
Liu S F. Grey system theory and its application[M]. Beijing: Science Press, 2021. | |
3 | 熊萍萍,陈诗婷,周依凡,等.基于新型核与灰度序列的MGM(1,m,N)模型及其应用[J].中国管理科学,2022,30(7):130-139. |
Xiong P P, Chen S T, Zhou Y F, et al. MGM (1,m,N) model based on the new kernel and degree of greyness sequences and lts application[J]. Chinese Journal of Management Science, 2022,30(7):130-139. | |
4 | 李惠,曾波,周文浩.基于灰色参数组合优化新模型的生活垃圾清运量预测研究[J].中国管理科学,2022,30(4):96-107. |
Li H, Zeng B, Zhou W H. Forecasting domestic waste clearing and transporting volume by employing a new grey parameter combination optimization model[J]. Chinese Journal of Management Science, 2022,30(4):96-107. | |
5 | 曾波,刘思峰,白云,等.基于灰色系统建模技术的人体疾病早期预测预警研究[J].中国管理科学,2020,28(1):144-152. |
Zeng B, Liu S F, Bai Y, et al. Grey system modeling technology for early prediction and warning of human diseases[J]. Chinese Journal of Management Science, 2020,28(1):144-152. | |
6 | Xie N M, Liu S F. Discrete grey forecasting model and its optimization[J]. Applied Mathematical Modeling, 2009, 33: 1173-1186. |
7 | 曾波,李树良,孟伟.灰色预测理论及其应用[M].北京:科学出版社,2020. |
Zeng B, Li S L, Meng W. Grey prediction theory and its application[M]. Beijing: Science Press, 2020. | |
8 | 张鹏.引入新息项的离散灰色预测模型[J].统计与决策,2020,36(5):29-32. |
Zhang P. A discrete grey prediction model with innovation term[J]. Statistics & Decision, 2020,36(5):29-32. | |
9 | 曾亮,骆世广.分数阶累加的离散GM(2,1)模型与应用[J].重庆师范大学学报(自然科学版),2021,38(5):73-80+2. |
Zeng L, Luo S G. A discrete GM(2,1)model with fractional-order accumulation and its application[J]. Journal of Chongqing Normal University(Natural Science),2021,38(5):73-80+2. | |
10 | 曹阳,梁爽,沈琴琴,等.阻尼累加离散GM(1,1)模型及其应用[J].控制与决策,2023,38(6):1687-1694. |
Cao Y, Liang S, Shen Q Q, et al. Damping accumulated discrete gm(1,1) model and its application[J]. Control and Decision,2023,38(6):1687-1694. | |
11 | 傅家俊,尹泉,傅鹤林,等.基于优化的灰色离散Verhulst新陈代谢模型的基坑沉降预测[J].公路工程,2019,44(2):19-22+120. |
Fu J J, Yin Q, Fu H L, et al. Settlement prediction of foundation pit based on optimized grey discrete verhulst metabolic model[J]. Highway Engineerin, 2019,44(2):19-22+120. | |
12 | Xie N M, Liu S F, Yang Y J, et al. On novel grey forecasting model based on non-homogeneous index sequence[J]. Applied Mathematical Modelling,2013,37(7): 5059-5068. |
13 | 谢乃明,刘思峰 .近似非齐次指数序列的离散灰色预测模型特性研究[J].系统工程与电子技术,2008,30(5):863-868. |
Xie N M, Liu S F. Research on the non-homogenous discrete grey model and its parameters properties[J]. Systems Engineering and Electronics,2008,30(5): 863-868. | |
14 | 曹邦兴.差分方程灰色DEGM(2,1)模型的优化及应用[J].统计与决策, 2020, 36(13):57-60. |
Cao B X. Forecast model of ecological economic warning degree based on gray theory and its applications[J]. Statistics & Decision, 2020, 36(13):57-60. | |
15 | 孟伟,曾波.基于互逆分数阶算子的离散灰色模型及阶数优化[J].控制与决策,2016,31(10):1903-1907. |
Meng W, Zeng B. Discrete grey model with inverse fractional operators and optimized order[J]. Control and Decision, 2016, 31(10):1903-1907. | |
16 | Gou X Y, Zeng B, Gong Y. An improved multi-variable grey model for forecasting China’s finished products from comprehensive waste utilization[J]. Environmental Science and Pollution Research, 2021, 28: 42901-42915. |
17 | 邹国焱,魏勇.广义离散灰色预测模型及其应用[J].系统工程理论与实践,2020,40(3):736-747. |
Zou G Y, Wei Y. Generalized discrete grey model and its application[J]. Systems Engineering-Theory & Practice, 2020, 40(3):736-747. | |
18 | 罗党,韦保磊.一类离散灰色预测模型的统一处理方法及应用[J].系统工程理论与实践,2019,39(2):451-462. |
Luo D, Wei B L. A unified treatment approach for a class of discrete grey forecasting models and its application[J]. Systems Engineering-Theory & Practice, 2019,39(2):451-462. | |
19 | 谢乃明,刘思峰.离散灰色模型的仿射特性研究[J].控制与决策,2008(2):200-203. |
Xie N M, Liu S F. Research on the affine properties of discrete grey model[J]. Control and Decision, 2008(2):200-203. | |
20 | 吴正朋,刘思峰,党耀国,等.再论离散GM(1,1)模型的病态问题研究[J].系统工程理论与实践,2011,31(1):108-112. |
Wu Z P, Liu S F, Dang Y G, et al. study on the morbidity problem in the grey model[J]. Systems Engineering-Theory & Practice, 2011, 31(1):108-112. | |
21 | Liu Y T, Yang Y, Pan F, et al. A conformable fractional unbiased grey model with a flexible structure and it’s application in hydroelectricity consumption prediction[J]. Journal of Cleaner Production, 2022, 367:133029. |
22 | Wei B L. Parameter estimation strategies for separable grey system models with comparisons and applications[J]. Applied Mathematical Modelling, 2023, 116: 32-44. |
23 | Wang Y, Yang Z S, Ye L L, et al. A novel self-adaptive fractional grey Euler model with dynamic accumulation order and its application in energy production prediction of China[J]. Energy, 2023, 265: 126384. |
24 | 吉培荣, 黄巍松, 胡翔勇. 无偏灰色预测模型[J]. 系统工程与电子技术, 2000(6): 6-7+80. |
Ji P R, Huang W S, Hu X Y. An unbiased grey forecasting model[J]. Systems Engineering and Electronics, 2000(6): 6-7+80. | |
25 | 刘晓梅,周钢.基于精细积分法的无偏非齐次灰色模型构建[J].控制与决策,2022,37(11):3058-3064. |
Liu X M, Zhou G. An unbiased non-homogeneous grey model based on high precise direct integration method[J]. Control and Decision, 2022,37(11):3058-3064. | |
26 | Zheng C L, Wu W Z, Xie W L, et al. Forecasting the hydroelectricity consumption of China by using a novel unbiased nonlinear grey Bernoulli model[J]. Journal of Cleaner Production, 2021, 278: 123903. |
27 | 白旻,张旻昱,王晓超.碳中和背景下全球新能源汽车产业发展政策与趋势[J].信息技术与标准化,2021(12):13-20. |
Bai M, Zhang M Y, Wang X C. Development policies and trends of global new energy vehicle industry under the background of carbon neutrality[J]. Information Technology & Standardization, 2021(12):13-20. | |
28 | 中国汽车技术研究中心,日产(中国)投资有限公司,东风汽车有限公司.中国新能源汽车产业发展报告(2021)[M].北京:社会科学文献出版社,2021. |
China Automotive Technology and Research Center Co., Ltd., Nissan Group of China Co., Ltd., DONGFENG Motor Company Limited. Annual report on new energy vehicle industry in China(2020) [M]. Beijing: Social Sciences Academic Press (CHINA), 2021. |
[1] | 刘连义,刘思峰,吴利丰. 基于离散时间灰色幂模型的新能源汽车销售量预测[J]. 中国管理科学, 2024, 32(1): 106-114. |
[2] | 周伟杰, 姜慧敏, 成雨珂, 党耀国, 丁松. 基于可重复性分数阶灰色时间幂模型的中国水电消费预测研究[J]. 中国管理科学, 2023, 31(5): 279-286. |
[3] | 曾波, 苟小义, 张志伟. 基于区间灰作用量的GM(1,1)均值差分模型解的非唯一性研究[J]. 中国管理科学, 2022, 30(10): 247-255. |
[4] | 熊萍萍, 陈诗婷, 周依凡, 刘煜淳, 丁松. 基于新型核与灰度序列的MGM(1,m,N)模型及其应用[J]. 中国管理科学, 2022, 30(7): 130-139. |
[5] | 李惠,曾波,周文浩. 基于灰色参数组合优化新模型的生活垃圾清运量预测研究[J]. 中国管理科学, 2022, 30(4): 96-107. |
[6] | 李晔, 丁圆苹. 基于驱动因素控制的线性时变参数DLDGM(1,N)模型[J]. 中国管理科学, 2022, 30(3): 221-229. |
[7] | 段辉明, 肖新平. 惯性灰色模型及其应用性质研究[J]. 中国管理科学, 2022, 30(3): 269-279. |
[8] | 夏杰, 马新, 吴文青. 基于Simpson公式改进的FAGM(1,1)模型及其应用研究[J]. 中国管理科学, 2021, 29(5): 240-248. |
[9] | 钱丽丽, 刘思峰, 邓桂丰. 考虑后悔规避的灰色群体偏离靶心度决策方法[J]. 中国管理科学, 2020, 28(6): 193-200. |
[10] | 曾波, 刘思峰, 白云, 周猛. 基于灰色系统建模技术的人体疾病早期预测预警研究[J]. 中国管理科学, 2020, 28(1): 144-152. |
[11] | 丁松, 李若瑾, 党耀国. 基于初始条件优化的GM(1,1)幂模型及其应用[J]. 中国管理科学, 2020, 28(1): 153-161. |
[12] | 吴利丰, 于亮, 文朝霞. 预测复杂装备研制费用的GM(0,N)模型[J]. 中国管理科学, 2019, 27(7): 203-207. |
[13] | 周伟杰, 张宏如, 党耀国, 王正新. 新息优先累加灰色离散模型的构建及应用[J]. 中国管理科学, 2017, 25(8): 140-148. |
[14] | 曾波, 刘思峰, 曲学鑫. 一种强兼容性的灰色通用预测模型及其性质研究[J]. 中国管理科学, 2017, 25(5): 150-156. |
[15] | 杨保华, 赵金帅. 优化离散灰色幂模型及其应用[J]. 中国管理科学, 2016, 24(2): 162-168. |
阅读次数 | ||||||
全文 |
|
|||||
摘要 |
|
|||||
|