美文网首页
FRM数量分析

FRM数量分析

作者: Agoni_1f59 | 来源:发表于2020-10-13 12:53 被阅读0次

Time Series:

Cycle,Seasonality,Trend

1. Stationary time series:Cycle(可预测)

cycle 一定可以 mean reversion

mean reversion 不一定 cycle


Covariance Stationary:

时间序列有预测性的前提

同时满足:

①mean:constant and finite

②variance:constant and finite

③autocovariance:在|tao|不变时,constant and finite


①Autocovariance function:

如果time series满足covariance stationary,那么它的autocovariance只取决于displacement(tao),而不取决于time(t)

②Autocorrelation function(ACF):

③Autoregression(AR):

the variable is regressed on lagged values of itself

④Partial autocorrelation function(PCF):

相关文章

网友评论

      本文标题:FRM数量分析

      本文链接:https://www.haomeiwen.com/subject/nolvpktx.html