STATISTICAL PREDICTABILITY OF CLIMATE TIME SERIES.

被引:0
|
作者
机构
来源
Sov Meteorol Hydrol | 1986年 / 6卷 / 6-13期
关键词
MATHEMATICAL MODELS - STATISTICAL METHODS - Time Series Analysis;
D O I
暂无
中图分类号
学科分类号
摘要
A theoretical study is made of first- and second-order autoregression/sliding mean (ARSM) models in all areas of their existence. Calculated a priori are normalized standards of forecasts with different lead times as a function of autocorrelations, which permits determination of the limits of predictability of climate time series. Approximations of such series by models of the signal plus noise type are discussed. Studied analytically are procedures for identifying the signal and constructing predictable climate statistics by means of space and time averaging of the implementations.
引用
收藏
相关论文
共 50 条