Timeline for Serial correlation
Current License: CC BY-SA 3.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 7, 2018 at 19:20 | comment | added | Sympa | Well exogenous variables do not typically resolve serial correlations. Lagged Y variables most often do. But, as you experience sometimes you have to add more than one lag. And, as mentioned such models become increasingly meaningless. I think when you add lags of Y, they have a mean reverting impact on the residuals. And, therefore it attenuates the serial correlation. | |
| Jan 7, 2018 at 3:13 | comment | added | Thu Hương | I know that both : lagged and strictly exogenous are 2 things that can correct the problem of serial correlation, but I'm not clear about the way they do. Please explain to me if you know it. Thank you! | |
| Jan 6, 2018 at 23:19 | comment | added | Sympa | I focused my answer on the specific question. | |
| Jan 6, 2018 at 19:18 | comment | added | IrishStat | Are you not assuming that the error process is free of deterministic structure such as Pulses, Level/step shifts , Seasonal Pulses,Local Time Trends AND that the error variance is homogeneous over time in order for what you have written to be true? | |
| Jan 6, 2018 at 19:09 | history | answered | Sympa | CC BY-SA 3.0 |