固体潮观测序列的AR建模方法

AR MODELING METHOD FOR OBSERVATION SERIES OF SOLID TIDE

证明了某一波群的固体潮理论值加上一个白噪声可用一个ＡＲ（２） 过程表示，固体潮观测序列可用高阶的ＡＲ模型逼近。给出了由ＡＲ 模型参数求解观测序列功率谱的方法，并用九江台资料作了实例分析。该方法可用于对短观测序列进行谱分析，以便跟踪地震短临前兆。

In this article, it is proved that the combination of the theoretical value of solid tide with white noise can be described by AR(2) model. The observation series of solid tide can be approached by using a high order AR model. A spectroanalysis method from AR model parameters is given. As an example, the tide data of Jiujiang seismostation are analyzed by AR model. This method can be used to make spectroanalysis for short series in order to trace the short term and immediate precursor of earthquake.\;