摘要
针对非等间距时间序列预测中存在误差较大的问题,本文建立了非等间距相关系数AR(p)序列预测方法,详细讨论了非等间距相关系数AR(1)序列和AR(2)序列的预测公式和误差估计。大量计算表明,本文方法与通过插值将非等间距序列变换为等间距序列进行预测的传统方法相比,具有更高的预测精度。
A new prediction method for correlation coefficient stationary series with unequally spaced data is established. The prediction formulas and error estimations for the correlation coefficient AR(1) and AR(2) series are discussed in detail. The traditional interpolation method, which transfers the unequally spaced data to equally spaced data, may cause a great error in prediction. The results of calculations show that the precision of present method is higher than the traditional interpolation method.
出处
《航空动力学报》
EI
CAS
CSCD
北大核心
2003年第6期713-716,共4页
Journal of Aerospace Power
基金
国防科技预研项目(413200204)
关键词
非等间距相关系数
误差估计
时间序列预测方法
预测精度
非平衡随机过程
aerospace propulsion system
time series
unequally spaced series
correlation coefficient stationary series
prediction
non-stationary random processes
