摘要
根据Ｂａｙｅｓ准则，提出一种时间序列与多元分析相结合的广义判别模式．该模式既保留了时间序列线性模式的优点，同时又吸收了多元分析线性模式的长处．从气象状况演变的物理机制来看，这种综合考虑系统状况的方法比单纯考虑前期状态演变或单纯考虑某一时次的状态更有意义．
A discriminant prediction model of time series and multivariate analysis is introduced.In this model,the dynamic and static characteristics of the forecast system are considered.According to the physical temporal mechanism of the weather states,the model considering the system states comprehensively is more effective to predict weather states.
出处
《云南大学学报（自然科学版）》
CAS
CSCD
1998年第1期71-74,共4页
Journal of Yunnan University(Natural Sciences Edition)
关键词
Bayes准则
广义判别模式
天气预报
统计预报
multivariate analysis,time series analysis,Bayes criterion,discriminant analysis,forecast
