基于子系统建模与组合预测的原油价格预测方法研究
|
摘要
石油在全球经济中举足轻重的战略地位决定了原油价格预测的必要性。当前世界经济正处于经济复苏的关键时期,中国的国际地位日益提升,这种特殊的国际环境决定了原油价格预测的紧迫性。本文以WTI原油价格为预测标的,将组合预测的三个层次:预测信息的组合,预测方法的组合以及预测结果的组合充分应用在油价预测的过程中。首先提炼出影响国际油价的五大主要因素——供给因素,需求因素,库存因素,经济因素,金融因素。将每个影响因素视为一个子系统。然后对各个子系统分别建立油价预测模型,在子系统建模过程中,先采用Granger因果检验筛选出对原油价格有预测意义的变量,在此因果关系模型的基础上加入自回归模型,这体现了组合预测的第二层次——预测方法的组合。在评价五个子系统的预测能力时,引入假设检验的思想,与传统的预测效果评价方法相比,该方法在得出判定孰优孰劣时更加慎重。最后对子系统的预测结果进行等权重组合,得到更加稳健的预测结果。在定量预测的同时,根据五个子系统中的景气指标变量构造―油价景气指数‖,实现了组合预测的第一个层次——预测信息的组合。本文实证研究表明,油价景气指数在预测油价下个月的走势时具有较高的准确性。本文将定量分析的结果与定性分析的结论相互融合,相对全面地展示了国际原油市场的动态。
The fact that crude oil holds an important and indispensable position in the global economy determines the necessity for crude oil price forecast. The world economy is now in a critical period of recovery and China's international status is also enhanced greatly, both of which determine the urgency of crude oil price forecast. Price of WTI is forecasted in this paper through the full application of three levels of Forecast Combination: combining forecast information, combining forecast methods and combining the forecast results. At first, five main factors influencing crude oil price are extracted, which are named supply factor, demand factor, stock factor, economy factor and finance factor. Each factor is considered as a subsystem. Then five forecast sub models are built in five subsystems respectively. In the subsystem modeling process, firstly the variables of top importance are filtered out by Granger Causality Test, and then two classical forecast methods (Causality Model and Autoregresstion Model) are combined to forecast the price of crude oil. On evaluating the forecast accuracy of five subsystems, hypothesis testing is originally occupied, which outperforms the traditional evaluation methods in the empirical analysis of this paper. Finally, the forecast results of these five subsystems are combined to act as the ultimate forecast result, which is more robust than any of those five ones. Apart from the quantitative analysis, a "Crude Oil Prosperity Index" is fabricated according to the performance of some certain variables in this paper, which is the combination of forecasting information. It is verified by empirical research that the Crude Oil Prosperity Index is doing well in predicting the price trend. The results of quantitative analysis and qualitative analysis are integrated to present a relatively full view of the international oil market.
The fact that crude oil holds an important and indispensable position in the global economy determines the necessity for crude oil price forecast. The world economy is now in a critical period of recovery and China's international status is also enhanced greatly, both of which determine the urgency of crude oil price forecast. Price of WTI is forecasted in this paper through the full application of three levels of Forecast Combination: combining forecast information, combining forecast methods and combining the forecast results. At first, five main factors influencing crude oil price are extracted, which are named supply factor, demand factor, stock factor, economy factor and finance factor. Each factor is considered as a subsystem. Then five forecast sub models are built in five subsystems respectively. In the subsystem modeling process, firstly the variables of top importance are filtered out by Granger Causality Test, and then two classical forecast methods (Causality Model and Autoregresstion Model) are combined to forecast the price of crude oil. On evaluating the forecast accuracy of five subsystems, hypothesis testing is originally occupied, which outperforms the traditional evaluation methods in the empirical analysis of this paper. Finally, the forecast results of these five subsystems are combined to act as the ultimate forecast result, which is more robust than any of those five ones. Apart from the quantitative analysis, a "Crude Oil Prosperity Index" is fabricated according to the performance of some certain variables in this paper, which is the combination of forecasting information. It is verified by empirical research that the Crude Oil Prosperity Index is doing well in predicting the price trend. The results of quantitative analysis and qualitative analysis are integrated to present a relatively full view of the international oil market.
引文
[1]张学武,杨祺, 2007年国际市场石油价格变化特点分析及后期走势预测[J],中国能源, 2007, 29(8), 25~29
[2]李继尊,姜克隽,当前高油价不会改变中国经济增长的基本趋势[J],中国能源, 2007, 29(6): 30~33
[3]舒通,基于变系数回归模型的石油价格预测[J],数理统计与管理, 2008, 27(5): 818~822
[4]丁静之,阂骥,林怡, ARIMA模型在石油价格预测中的应用[J],物流技术, 2008, 27(10): 156~160
[5]冯春山,吴家春,蒋馥,石油价格的组合预测研究[J],石油大学学报, 2004 , 20(1): 12~14
[6]郑恺,谷耀,国内油价高涨的主因——基于VAR的实证分析[J],南方经济, 2006, 5: 84~90
[7]李红军,范玉妹,程蕾,用带时序子模块的系统动力学模型预测Brent原油价格[J],应用基础与工程科学学报, 2003 , 11(2 ): 217~220
[8]林敏,刘志斌,油价系统的随机模拟[J],系统工程, 2008 , 26(2): 73～77
[9]梁强,范英,魏一鸣,基于小波分析的石油价格长期趋势预测方法及其实证研究[J],中国管理科学, 2005, 13(1): 32~38
[10]刘孝成,基于分形理论的石油价格预测[J],西安石油大学学报, 2009, 3:19~22
[11] Ying Fan , Qiang Liang , Yi-Ming Wei ,A generalized pattern matching approach for multi-step prediction of crude oil price[J], Energy Economics , 2008, 30: 889~904
[12] S,K, Aggarwal, L,M, Saini , Ashwani Kumar, Price forecasting using wavelet transform and LSE based mixed model in Australian electricity market[J], International Journal of Energy SectorManagement, 2008, 2(4): 521~546
[13] Viviana Fernandez, Traditional versus Novel Forecasting Techniques: How Much do We Gain? [J], Journal of Forecasting, 2008, 27: 637~648
[14]汪同三,张涛等,组合预测——理论,方法及应用[M ],北京:社会科学文献出版社, 2008, 5~102
[15]高铁梅,计量经济分析方法与建模EViews应用及实例[M],北京:清华大学出版社, 2006, 267~314
[16] O, Felix Ayadi , Johnnie Williams, Ladelle M, Hyman, Fractional dynamic behavior in Forcados Oil Price Series: An application of detrended fluctuation analysis[J], Energy for Sustainable Development, 2009,13: 11~17
[17] Ray Bachan, Oil demand in China:an econometric approach [J], International Journal of Emerging Markets, 2008, 3(1): 54~70
[18]钟德寿,修晶,袁霓,国际石油价格波动与中国的应对措施——基于博弈论与ARCH类模型的分析[J],中国青年政治学院学报, 2008,3: 26~29
[19]赵振智,胡昌雷,国际石油价格及趋势预测[J],中国能源, 2007, 29(8): 29~35
[20]陈彦玲,国际油价上涨对中美居民消费价格指数的影响[J],中国流通经济, 2007, 9: 50~52
[21]胡国松,冯雪梅,金融危机下的国际石油市场走势与我国对策[J],中外能源, 2009,1: 12~15
[22] Dr Salman Saif Ghouri, Forecasting natural gas prices using cointegration technique[J], OPEC Review 2006, l 6: 249~269
[23]程伟力,影响国际油价因素的定量分析[J],国际石油经济, 2005, 13 (8): 15~18
[24] Jean-thomas bernard,Llynda khalaf, Forecasting Commodity Prices:GARCH, Jumps, and Mean Reversion[J], Journal of Forecasting, 2008, 27: 279~291
[25]李海英,唐衍伟,罗婷,中国燃料油价格国际市场相关性的实证研究[J],资源科学, 2009, 1(1): 30~32
[26]韩冬炎,陈蕊,对石油价格走势预测的数理研究——基于分形方法的应用[J],价格理论与实践, 2007, 2: 12~14
[27]程雪婷,齐中英,国际石油价格波动中的奇异吸引子分析[J],管理工程学报, 2005, 2:11~13
[28] Albert wang, A study of the oil futures market[J], The Journal of Futures Markets, 2009, 29(5): 414~429
[29]冯春山,吴家春,石油价格的ARFIMA模型预测研究[J],上海理工大学学报, 2005,27(6):33~35
[30]王延明,用数理方法预测石油价格[J],北京石油管理干部学院学报, 2005, 6: 4~7
[31] Thomas a, knetsch , Forecasting the Price of Crude Oil via Convenience Yield Predictions[J], Journal of Forecasting, 2007, 26: 527~549
[32]刘立霞,马军海,基于一种变权重组合的石油期货价格预测研究[J],计算机工程与应用, 2008, 44: 230~231
[33]赵晓燕,徐克龙,石油价格的一种变权重组合预测[J],统计与决策, 2006, 9: 54~56
[34]周四清,王坚强,基于多准则优化的组合预测方法[J],系统工程与电子技术, 2009, 31(7): 1651~1654
[35]彭斯俊,朱四如,基于非单一类型的组合模型在回归分析中的应用[J],数学杂志, 2009, 29(4): 77~79
[36]周超,王秀芝,组合预测在油气操作成本预测中的应用[J],西南石油大学学报, 2009, 2(4): 28~32
[37]郑建湖,基于最优加权的道路交通事故组合预测模型[J],中国安全科学学报, 2009, 1 9(7): 44~47
[38]高扬,刘严崴,基于支持向量回归机的油田系统预测方法[J]大庆石油学院学报, 2007,31(1): 93~96
[39]周向彤,近期国际石油价格回顾与预测[J],国际石油经济, 2009, 2: 79~81
[40] Sathya Swaroop Debasish, Effect of futures trading on spot-price volatility: evidence for NSE Nifty using GARCH[J], The Journal of Risk Finance, 2009, 10(1): 67~77
[41]马超群,李科,基于协整和GRACH模型分析——中国油价波动特征[J],求索, 2004, 12: 8~10
[42]余炜彬,范英,魏一鸣,焦建玲, Brent原油期货市场的协整性分析[J],数理统计与管理, 2004, 5(23): 28~32
[43]李畅,杨再斌,国际石油价格波动特点及影响因素的实证分析[J],资源科学, 2007, 29 (1): 42~45
[44]李晓明,国际石油价格之残差自回归模型短期预测[J],统计与决策, 2008, 22: 144~147
[45]高辉,国内外燃料油价格关联度及动态滚动预测的模型研究——基于日数据的实证分析[J],国际石油经济, 2005, 12(13): 24~28,
[46] Evan J, McSweeney, A comparative analysis of oil as a risk factor in Australian industry stock returns, Studies in Economics and Finance, 2008, 25(2): 131~145
[47] Javad Kargar , Oil and Currency Factors in Middle East Equity Returns[J], Managerial Finance , 2004, 30(3): 3~16
[48]宋芳,向小东,径向基函数网络的改进算法在油价预测中的应用[J],科技与产业, 2007, 7(1): 40~44
[49]徐海燕,石油价格波动期中的油价预测模型[J],国际石油经济, 2005, 13 (1): 50~53
[50]李捷,候秀红,韩志杰,基于卡尔曼滤波和小波的网络流量预测算法研究[J],电子与信息学报, 2007, 29(3): 726~729
[51]孙大利,基于神经网络的石油价格预测[J],石油化工高等学校学报2006, l9(2): 90~93
[52]马社祥,刘贵忠,曾召华,基于小波分析的非平稳时间序列分析与预测[J],系统工程学报, 2000, 15(4): 307~310
[53] Ali Ghaffari, Samaneh Zare, A novel algorithm for prediction of crude oil price variation based on soft computing[J], Energy Economics , 2009, 31: 531~536
[54]邓凯旭宋宝瑞,小波变换在金融数据分析中的应用[J],数理统计与管理, 2006, 25(2): 206~210
[55]周景宏,郗伟东,禹海兰,小波方法在股市分析中的应用[J],东北电力大学学报, 2006, 26(6): 48~52
[56]向小东,小波神经网络预测方法在石油期货价格预测中的应用研究[J],技术经济, 2006, 25(6): 121~125
[57]Mohammad R, Jahan~Parvar ,Oil prices and competitiveness: time series evidence from six oil~producing countries [J], Journal of Economic Studies, 2009, 36(1): 98~118
[2]李继尊,姜克隽,当前高油价不会改变中国经济增长的基本趋势[J],中国能源, 2007, 29(6): 30~33
[3]舒通,基于变系数回归模型的石油价格预测[J],数理统计与管理, 2008, 27(5): 818~822
[4]丁静之,阂骥,林怡, ARIMA模型在石油价格预测中的应用[J],物流技术, 2008, 27(10): 156~160
[5]冯春山,吴家春,蒋馥,石油价格的组合预测研究[J],石油大学学报, 2004 , 20(1): 12~14
[6]郑恺,谷耀,国内油价高涨的主因——基于VAR的实证分析[J],南方经济, 2006, 5: 84~90
[7]李红军,范玉妹,程蕾,用带时序子模块的系统动力学模型预测Brent原油价格[J],应用基础与工程科学学报, 2003 , 11(2 ): 217~220
[8]林敏,刘志斌,油价系统的随机模拟[J],系统工程, 2008 , 26(2): 73～77
[9]梁强,范英,魏一鸣,基于小波分析的石油价格长期趋势预测方法及其实证研究[J],中国管理科学, 2005, 13(1): 32~38
[10]刘孝成,基于分形理论的石油价格预测[J],西安石油大学学报, 2009, 3:19~22
[11] Ying Fan , Qiang Liang , Yi-Ming Wei ,A generalized pattern matching approach for multi-step prediction of crude oil price[J], Energy Economics , 2008, 30: 889~904
[12] S,K, Aggarwal, L,M, Saini , Ashwani Kumar, Price forecasting using wavelet transform and LSE based mixed model in Australian electricity market[J], International Journal of Energy SectorManagement, 2008, 2(4): 521~546
[13] Viviana Fernandez, Traditional versus Novel Forecasting Techniques: How Much do We Gain? [J], Journal of Forecasting, 2008, 27: 637~648
[14]汪同三,张涛等,组合预测——理论,方法及应用[M ],北京:社会科学文献出版社, 2008, 5~102
[15]高铁梅,计量经济分析方法与建模EViews应用及实例[M],北京:清华大学出版社, 2006, 267~314
[16] O, Felix Ayadi , Johnnie Williams, Ladelle M, Hyman, Fractional dynamic behavior in Forcados Oil Price Series: An application of detrended fluctuation analysis[J], Energy for Sustainable Development, 2009,13: 11~17
[17] Ray Bachan, Oil demand in China:an econometric approach [J], International Journal of Emerging Markets, 2008, 3(1): 54~70
[18]钟德寿,修晶,袁霓,国际石油价格波动与中国的应对措施——基于博弈论与ARCH类模型的分析[J],中国青年政治学院学报, 2008,3: 26~29
[19]赵振智,胡昌雷,国际石油价格及趋势预测[J],中国能源, 2007, 29(8): 29~35
[20]陈彦玲,国际油价上涨对中美居民消费价格指数的影响[J],中国流通经济, 2007, 9: 50~52
[21]胡国松,冯雪梅,金融危机下的国际石油市场走势与我国对策[J],中外能源, 2009,1: 12~15
[22] Dr Salman Saif Ghouri, Forecasting natural gas prices using cointegration technique[J], OPEC Review 2006, l 6: 249~269
[23]程伟力,影响国际油价因素的定量分析[J],国际石油经济, 2005, 13 (8): 15~18
[24] Jean-thomas bernard,Llynda khalaf, Forecasting Commodity Prices:GARCH, Jumps, and Mean Reversion[J], Journal of Forecasting, 2008, 27: 279~291
[25]李海英,唐衍伟,罗婷,中国燃料油价格国际市场相关性的实证研究[J],资源科学, 2009, 1(1): 30~32
[26]韩冬炎,陈蕊,对石油价格走势预测的数理研究——基于分形方法的应用[J],价格理论与实践, 2007, 2: 12~14
[27]程雪婷,齐中英,国际石油价格波动中的奇异吸引子分析[J],管理工程学报, 2005, 2:11~13
[28] Albert wang, A study of the oil futures market[J], The Journal of Futures Markets, 2009, 29(5): 414~429
[29]冯春山,吴家春,石油价格的ARFIMA模型预测研究[J],上海理工大学学报, 2005,27(6):33~35
[30]王延明,用数理方法预测石油价格[J],北京石油管理干部学院学报, 2005, 6: 4~7
[31] Thomas a, knetsch , Forecasting the Price of Crude Oil via Convenience Yield Predictions[J], Journal of Forecasting, 2007, 26: 527~549
[32]刘立霞,马军海,基于一种变权重组合的石油期货价格预测研究[J],计算机工程与应用, 2008, 44: 230~231
[33]赵晓燕,徐克龙,石油价格的一种变权重组合预测[J],统计与决策, 2006, 9: 54~56
[34]周四清,王坚强,基于多准则优化的组合预测方法[J],系统工程与电子技术, 2009, 31(7): 1651~1654
[35]彭斯俊,朱四如,基于非单一类型的组合模型在回归分析中的应用[J],数学杂志, 2009, 29(4): 77~79
[36]周超,王秀芝,组合预测在油气操作成本预测中的应用[J],西南石油大学学报, 2009, 2(4): 28~32
[37]郑建湖,基于最优加权的道路交通事故组合预测模型[J],中国安全科学学报, 2009, 1 9(7): 44~47
[38]高扬,刘严崴,基于支持向量回归机的油田系统预测方法[J]大庆石油学院学报, 2007,31(1): 93~96
[39]周向彤,近期国际石油价格回顾与预测[J],国际石油经济, 2009, 2: 79~81
[40] Sathya Swaroop Debasish, Effect of futures trading on spot-price volatility: evidence for NSE Nifty using GARCH[J], The Journal of Risk Finance, 2009, 10(1): 67~77
[41]马超群,李科,基于协整和GRACH模型分析——中国油价波动特征[J],求索, 2004, 12: 8~10
[42]余炜彬,范英,魏一鸣,焦建玲, Brent原油期货市场的协整性分析[J],数理统计与管理, 2004, 5(23): 28~32
[43]李畅,杨再斌,国际石油价格波动特点及影响因素的实证分析[J],资源科学, 2007, 29 (1): 42~45
[44]李晓明,国际石油价格之残差自回归模型短期预测[J],统计与决策, 2008, 22: 144~147
[45]高辉,国内外燃料油价格关联度及动态滚动预测的模型研究——基于日数据的实证分析[J],国际石油经济, 2005, 12(13): 24~28,
[46] Evan J, McSweeney, A comparative analysis of oil as a risk factor in Australian industry stock returns, Studies in Economics and Finance, 2008, 25(2): 131~145
[47] Javad Kargar , Oil and Currency Factors in Middle East Equity Returns[J], Managerial Finance , 2004, 30(3): 3~16
[48]宋芳,向小东,径向基函数网络的改进算法在油价预测中的应用[J],科技与产业, 2007, 7(1): 40~44
[49]徐海燕,石油价格波动期中的油价预测模型[J],国际石油经济, 2005, 13 (1): 50~53
[50]李捷,候秀红,韩志杰,基于卡尔曼滤波和小波的网络流量预测算法研究[J],电子与信息学报, 2007, 29(3): 726~729
[51]孙大利,基于神经网络的石油价格预测[J],石油化工高等学校学报2006, l9(2): 90~93
[52]马社祥,刘贵忠,曾召华,基于小波分析的非平稳时间序列分析与预测[J],系统工程学报, 2000, 15(4): 307~310
[53] Ali Ghaffari, Samaneh Zare, A novel algorithm for prediction of crude oil price variation based on soft computing[J], Energy Economics , 2009, 31: 531~536
[54]邓凯旭宋宝瑞,小波变换在金融数据分析中的应用[J],数理统计与管理, 2006, 25(2): 206~210
[55]周景宏,郗伟东,禹海兰,小波方法在股市分析中的应用[J],东北电力大学学报, 2006, 26(6): 48~52
[56]向小东,小波神经网络预测方法在石油期货价格预测中的应用研究[J],技术经济, 2006, 25(6): 121~125
[57]Mohammad R, Jahan~Parvar ,Oil prices and competitiveness: time series evidence from six oil~producing countries [J], Journal of Economic Studies, 2009, 36(1): 98~118