尹同学2018-03-22 19:15:08
协整和线性回归有什么关系么
回答(1)
180****86962025-02-28 20:57:24
Cointegration and linear regression are related concepts, but they address different aspects of time series analysis. Cointegration refers to the statistical relationship between two or more non-stationary time series that are individually non-stationary but have a stable, long-term equilibrium relationship when combined. In other words, while each series may wander randomly, their linear combination may be stationary. On the other hand, linear regression typically models the relationship between two variables assuming that one is dependent on the other and that both are stationary, or that any trends in the data are removed. The connection lies in the fact that cointegration can often be detected through a form of linear regression, specifically by testing whether the residuals from a regression of one time series on another are stationary. If the residuals are stationary, the series are said to be cointegrated, indicating a meaningful long-term relationship between them.
- 评论(0)
- 追问(0)


评论
0/1000
追答
0/1000
+上传图片