基于区间时序模型的船舶价格指数预测

doi:10.6040/j.issn.1671-9352.4.2022.205

摘要/Abstract

摘要：

将远期运费协议(forward freight agreements, FFA)作为外生变量，研究其对船舶价格指数的具体影响。借助中点和极差方法, 分别建立区间型自回归模型和考虑区间型时间序列上、下限协整关系的区间型误差修正模型及带有外生变量FFA的区间型误差修正模型。将构建的模型用于对新造干散货船价格指数、二手干散货船价格指数进行的区间预测, 在平均绝对误差(MAE)、均方根误差(RMSE)指标上, 模型中加入协整项和FFA后预测精度更高。

关键词: 区间型时间序列, 船舶价格指数, 远期运费协议

Abstract:

This study considers the forward freight agreement (FFA) as an exogenous variable to analyze its specific impact on the vessel price index. With the center and range method, an interval autoregression model, an interval error correction model considering the co-integration between the upper and lower bounds of the interval time series, and interval error correction model with an additional incorporation of the exogenous variable FFA are established, respectively. The constructed models are employed to perform the interval prediction of the bulk carrier newship price index and bulk carrier secondhand price index. Based on the criteria MAE and RMSE, the prediction accuracy is higher after adding the co-integration term and FFA into the models.

Key words: interval time series, vessel price index, forward freight agreements

中图分类号:

F551

徐圆圆,郭红月,王利东. 基于区间时序模型的船舶价格指数预测[J]. 《山东大学学报(理学版)》, 2024, 59(1): 115-123.

Yuanyuan XU,Hongyue GUO,Lidong WANG. Interval time series models-based vessel price index forecasting[J]. JOURNAL OF SHANDONG UNIVERSITY(NATURAL SCIENCE), 2024, 59(1): 115-123.

图/表 12

表1

图1

图2

图3

表2

表3

图4

图5

表4

表5

表6

表7

参考文献 21

1	陈继红, 刘雯. 基于协整和VEC模型的VLCC新造船舶价格波动性[J]. 数学的实践与认识, 2016, 46 (24): 69- 78.
	CHEN Jihong , LIU Wen . Fluctuation of the price of newbuilding VLCC ships based on cointegration and VEC model[J]. Mathematics in Practice and Theory, 2016, 46 (24): 69- 78.
2	HER M T , CHUNG C C . Ship price predictions of panamax second-hand bulk carriers using grey models[J]. Journal of Marine Science and Technology, 2019, 27 (3): 229- 234.
3	TGEODORE S , MICHAEL T . Support vector machine algorithms: an application to ship price forecasting[J]. Computational Economics, 2021, 57, 55- 87. doi: 10.1007/s10614-020-10032-2
4	SYRIOPOULOS T , TSATSARONIS M , KARAMANOS I . Support vector machine algorithms: an application to ship price forecasting[J]. Computational Economics, 2021, 57, 55- 87. doi: 10.1007/s10614-020-10032-2
5	GAO R , LIU J H , BAI X W , et al. Annual dilated convolution neural network for newbuilding ship prices forecasting[J]. Neural Computing and Applications, 2022, 34, 11853- 11863. doi: 10.1007/s00521-022-07075-x
6	GAO R B , LIU J H , DU L , et al. Shipping market forecasting by forecast combination mechanism[J]. Maritime Policy and Management, 2022, 49 (8): 1059- 1074. doi: 10.1080/03088839.2021.1945698
7	WANG X , ZHANG Z , LI S . Set-valued and interval-valued stationary time series[J]. Journal of Multivariate Analysis, 2016, 145, 208- 223. doi: 10.1016/j.jmva.2015.12.010
8	SUN Y Y , HAN A , HONG Y M , et al. Threshold autoregressive models for interval valued time series data[J]. Journal of Econometrics, 2018, 206 (2): 414- 446. doi: 10.1016/j.jeconom.2018.06.009
9	MOORE R E. Interval analysis[D]. Englewood Cliffs: Prentice-Hall, 1966.
10	LIMA N , DE CARVALHO F A . An exponential-type kernel robust regression model for interval-valued variables[J]. Information Sciences, 2018, 2018 (454/455): 419- 442.
11	BILLARD L, DIDAY E. Regression analysis for interval-valued data & data analysis, classification and related methods[C]//Proceedings of the Seventh Conference of the International Federation of Classification Societies. Belgium: Springer, 2000: 369-374.
12	BILLARD L , DIDAY E . From the statistics of data to the statistics of knowledge: symbolic data analysis[J]. Journal of the American Statistical Association, 2003, 98 (462): 470- 487. doi: 10.1198/016214503000242
13	NETO L E A , CARVALHO F A T . Center and range method for fitting a linear regression model to symbolic interval data[J]. Computational Statistics and Data Analysis, 2008, 52 (3): 1500- 1515. doi: 10.1016/j.csda.2007.04.014
14	NETO L E A , CARVALHO F A T . Constrained linear regression models for symbolic interval-valued variables[J]. Computational Statistics and Data Analysis, 2010, 54 (2): 333- 347. doi: 10.1016/j.csda.2009.08.010
15	HAN A , HONG Y M , WANG S Y , et al. A vector autoregressive moving average model for interval-valued time series data[J]. Advances in Econometrics, 2016, 36, 417- 460.
16	YANG W , HAN A , WANG S Y . Analysis of crisis impact on crude oil prices: a new approach with interval time series modelling[J]. Quantitative Finance, 2016, 16 (12): 1917- 1928. doi: 10.1080/14697688.2016.1211795
17	杨威, 韩艾, 汪寿阳. 基于区间型数据的金融时间序列预测研究[J]. 系统工程学报, 2016, 31 (6): 817- 831.
	YANG Wei , HAN Ai , WANG Shouyang . Research on financial time Series prediction based on interval data[J]. Journal of Systems Engineering, 2016, 31 (6): 817- 831.
18	陈炜, 徐慧琳, 汪寿阳, 等. 基于误差修正与分解的区间值股价时间序列预测研究[J]. 系统工程理论与实践, 2023, 43 (2): 383- 397.
	CHEN Wei , XU Huilin , WANG Shouyang , et al. Error correction and decomposition method for forecast of interval-valued stock price time series[J]. Systems Engineering-Theory and Practice, 2023, 43 (2): 383- 397.
19	闵德权, 张鑫, 崔琪. 远期运费协议与新造船市场的波动溢出效应研究[J]. 中国水运, 2019, 19 (12): 37- 39.
	MIN Dequan , ZHANG Xin , CUI Qi . A study on volatility spillover effects of forward freight agreements and new shipbuilding market[J]. China Water Transport, 2019, 19 (12): 37- 39.
20	杨芊, 林国龙, 丁一. FFA市场与二手船市场的波动溢出效应[J]. 武汉理工大学学报(交通科学与工程版), 2015, 39 (5): 1001- 1004. doi: 10.3963/j.issn.2095-3844.2015.05.022
	YANG Qian , LIN Guolong , DING Yi . Volatility spillover effects between forward freight market and second-hand ship maket[J]. Journal of Wuhan University of Technology (Transportation Science and Engineering), 2015, 39 (5): 1001- 1004. doi: 10.3963/j.issn.2095-3844.2015.05.022
21	陶志富, 刘金培, 朱家明, 等. 区间值时间序列预测效果测度研究[J]. 模糊系统与数学, 2018, 32 (4): 135- 144.
	TAO Zhifu , LIU Jinpei , ZHU Jiaming , et al. Study on measurement of interval-value time series prediction effect[J]. Fuzzy Systems and Mathematics, 2018, 32 (4): 135- 144.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed

月份	新造干散货船		月份	二手干散货船
月份	Y_t^L	Y_t^U	月份	Y_t^L	Y_t^U
2000-01	107.43	107.87	2010-07	214.67	215.25
2000-02	108.10	108.60	2010-08	214.77	214.77
2000-03	108.73	108.85	2010-09	216.31	218.94
2000-04	108.80	109.42	2010-10	216.18	219.08
…	…	…	…	…	…

船类	变量	均值	中值	最大值	最小值	标准差	偏度	峰度	JB统计量	P值
新造船	Y_t^L	136.55	130.94	191.21	105.23	20.71	0.80	0.03	28.68	0.00
	Y_t^U	137.19	131.40	191.58	105.36	20.90	0.78	-0.02	26.97	0.00
	YC_t	136.87	131.19	191.40	105.35	20.80	0.79	0.00	27.80	0.00
	YR_t	0.33	0.19	5.73	0.01	0.57	5.71	43.26	22 374.49	0.00
	Δln YC_t	0.14	0.08	4.42	-6.13	1.25	-0.36	4.91	276.86	0.00
	Δln YR_t	0.00	0.00	4.05	-3.29	1.26	0.10	0.07	0.53	0.76
二手船	Y_t^L	126.18	118.35	216.31	70.64	34.93	0.75	0.05	13.10	0.00
	Y_t^U	129.15	119.15	219.08	71.15	35.89	0.71	-0.09	12.03	0.00
	YC_t	127.67	118.66	217.63	70.90	35.38	0.73	-0.02	12.54	0.00
	YR_t	1.68	1.25	9.02	0.05	1.64	1.60	2.94	112.55	0.00
	Δln YC_t	-0.18	-0.26	10.91	-10.03	3.54	0.03	0.48	1.66	0.44
	Δln YR_t	0.02	0.10	3.10	-3.49	1.37	-0.08	-0.17	0.25	0.88

船类	变量	lag	ADF	P值	结论	船类	变量	lag	ADF	P值	结论
新造船	ln Y_t^U	0	1.76	0.980	不平稳	二手船	lnY_t^U	0	-0.63	0.46	不平稳
	ln Y_t^U	1	0.77	0.865	不平稳		lnY_t^U	1	-0.48	0.51	不平稳
	lnY_t^L	0	1.48	0.964	不平稳		lnY_t^L	0	-0.63	0.45	不平稳
	lnY_t^L	1	1.05	0.921	不平稳		lnY_t^L	1	-0.41	0.53	不平稳
	ΔlnY_t^U	0	-7.70	≤0.01	平稳		Δln Y_t^U	0	-6.85	≤0.01	平稳
	ΔlnY_t^U	1	-5.30	≤0.01	平稳		Δln Y_t^U	1	-4.34	≤0.01	平稳
	lnY_t^L	0	-11.78	≤0.01	平稳		ln Y_t^L	0	-6.63	≤0.01	平稳
	lnY_t^L	1	-6.53	≤0.01	平稳		ln Y_t^L	1	-4.63	≤0.01	平稳
	u_t	0	-12.54	≤0.01	平稳		u_t	0	-8.99	≤0.01	平稳
	u_t	1	-8.53	≤0.01	平稳		u_t	1	-6.39	≤0.01	平稳

方程	变量	AR		ECM		ECM-X
方程	变量	系数	P值	系数	P值	系数	P值
中点方程	β₀^c	0.032 8	0.313 8	0.033 7	0.307 8	0.034 1	0.304 5
	β₁^c	0.549 1^***	0.000 0	0.541 8^***	0.000 0	0.533 8^***	0.000 0
	γ^c			-0.194 0^***	0.013 4	-0.215 8^***	0.007 1
	δ₁^c					0.649 8^*	0.024 6
	β₀^r	-0.014 7	0.416 8	-0.014 9	0.413 4	-0.014 9	0.413 6
极差方程	β₁^r	-0.510 6^***	0.000 0	-0.578 8^***	0.000 0	-0.578 8^***	0.000 0
	γ^r			-0.343 2^***	0.000 2	-0.343 1^***	0.000 1
	δ₁^r					-0.000 4	0.498 4

方程	变量	AR		ECM		ECM-X
方程	变量	系数	P值	系数	P值	系数	P值
中点方程	β₀^c	-0.234 2	0.181 1	-0.217 5	0.198 7	-0.216 4	0.186 7
	β₁^c	0.613 6^***	0.000 6	0.643 1^***	0.000 1	0.607 6^***	0.000 1
	γ^c			0.128 6	0.132 2	0.159 1	0.073 2
	δ₁^c					4.472 8^***	0.000 1
极差方程	β₀^r	0.028 9	0.405 2	0.032 3	0.386 9	0.032 8	0.385 6
	β₁^r	-0.461 4^***	0.000 1	-0.659 7^***	0.000 0	-0.660 3^***	0.000 0
	γ^r			-0.238 8^***	0.000 1	-0.238 9^***	0.000 1
	δ₁^r					0.099 6	0.259 4