| 赵庆.基于非线性组合模型对石油价格的预测[J].唐山学院学报,2014,27(6): |
| 基于非线性组合模型对石油价格的预测 |
| Oil Price Forecast Based on the Non-linear Combination Model |
| |
| DOI： |
| 中文关键词: 石油价格预测 HP 滤波 AR 模型 ARMA 模型 |
| 英文关键词: oil price forecast HP filter AR model ARMA model |
| 基金项目: |
|
| 摘要点击次数: 7257 |
| 全文下载次数: 6156 |
| 中文摘要: |
| 鉴于石油在国民经济中的重要地位及其对金融市场的重要影响，提出新的非线性组合预测模型：HP 滤波-AR 模型-ARMA 模型，使用 HP 滤波将原序列分解成两部分即周期性波动序列和趋势要素序列。根据两个序列的不同性质，对趋势要素序列建立 AR 模型，然后建立随机周期性波动序列，对随机周期性波动序列建立 ARMA 模型。将两个预测值之和与原序列比较，同时将该模型与其他模型进行对比。对比结果表明，预测精度高于其他模型。 |
| 英文摘要: |
| In view of the important role of oil in the national economy and its important effect on the financial markets,the author has developed a new non-linear combination forecasting model:HP filter + AR model + ARMA model.The HP filter divides the original sequence into two parts:periodic fluctuation sequence and trend factor sequence,according to different properties of which AR model is set up for the rend factor sequence,random periodic fluctuation sequence is es-tablished,and then ARMA model is set up for the random periodic fluctuation sequence.The sum of two predicted values is compared with the original sequence and the combination model is compared with other models.The comparison results show that the prediction accuracy of the combination model is higher than other models. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |