最大熵谱的优良特性和预报误差过滤系数阶数的确定
THE ADVANTAGE OF MAXIMUM ENTROPY METHOD AND DETERMINATION OF THE ORDER OF ITS PREDICTION ERROR COEFFICIENT
-
摘要: 最大熵谱分析比通常谱分析分辨率高,峰值偏离小,特别适宜于短资料或长周期的谱分析,这对气候分析和中长期过程的周期分析是有显著意义的。最大熵谱分析的预报误差过滤系数的阶数,可以在最终预报误差(FPE)m的前后两个极小值的谱估计一致的条件下,用前一个极小值的m数来确定。Abstract: The advantage of maximum entropy method(MEM) of spectral analysis is that it gives finer resolutions and less peak discrepomcy than the conventional spectral analysis. MEM is particularly useful for short data lengths and long period spectral analysis. It bears special signi ficance for periodic analysis of climate and medium-range and long-range process. The order of the coefficient of MEM prediction error may be determined by the order of the former minimum under the condition that the power of the former minimal final prediction error of spectral estimate is consistent with the latter.