CAJ | 학술논문

基于BDS法和替代数据法,对布伦特原油现货价、大庆原油现货价以及NYMEX天然气期货价和亨利中心天然气现货价的日数据进行了非线性特征检验.研究表明,上述能源价格收益率不服从正态分布,存在长相关特征并具有一定的非线性结构.在替代数据法检验时发现,嵌入维数小于6时,上述能源价格的原始数据和替代数据没有显著性差异,表明在维数低于6时,可以用少于6个经济变量进行线性模拟,以对相关能源价格进行短期预测,以便为中国能源政策制定及能源行业的科学发展提供有益启示.
기우BDS법화체대수거법,대포륜특원유현화개、대경원유현화개이급NYMEX천연기기화개화형리중심천연기현화개적일수거진행료비선성특정검험.연구표명,상술능원개격수익솔불복종정태분포,존재장상관특정병구유일정적비선성결구.재체대수거법검험시발현,감입유수소우6시,상술능원개격적원시수거화체대수거몰유현저성차이,표명재유수저우6시,가이용소우6개경제변량진행선성모의,이대상관능원개격진행단기예측,이편위중국능원정책제정급능원행업적과학발전제공유익계시.
Based on the BDS method and the surrogate data method, this paper studies the nonlinear characteristics of the Brent Oilspot price, Daqing Oil price, NYMEX Gas price and Henry Gas price. The results show that the return series of the above energy pricedata do not follow a normal distribution and have “ fat-tailedness” characteristics, the non-linear relationship and some non-linearstructure. In the Surrogate data test, the results found that when the embedding dimension is less than 6, there is no significantdifference among the original data and surrogate data of the above energy prices, which means that under the embedding dimension ofless than 6 the study may linearly simulate the energy prices with less than 6 economic variables in order to make the short-term forecastof the energy prices.