Prediction method and device for occurrence time of short-wave signal

The invention discloses a short wave signal appearance time prediction method and device, and belongs to the technical field of time sequence prediction. The prediction method comprises a VAR calculation step, a KDE calculation step and a prediction step. According to the method, the multi-dimensional feature vector is constructed based on time mapping, and meanwhile, the probability score capable of representing whether a certain time point appears or not is obtained by combining the results of KDE and VAR, so that the problem that the result deviates from historical behaviors due to the fact that the method is not linked with new data or excessively depends on the new data in the prior art is solved; a more accurate prediction result can be given based on historical behaviors while new data are considered. 本发明公开了一种短波信号出现时间的预测方法及装置，属于时间序列预测技术领域。所述预测方法包括包括VAR计算步骤、KDE计算步骤和预测步骤。本发明基于时间映射构造多维特征向量，同时结合KDE与VAR的结果获得可表征某时间点是否出现的概率评分，解决了现有技术中不与新数据联动或者过分依赖新数据导致结果偏离历史行为的问题，能够在兼顾新数据的同时基于历史行为给出更加准确地预测结果。