Timeline for Can second order differences stationary series be estimated via VECM model without considering support from economic theory?
Current License: CC BY-SA 4.0
13 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Feb 18, 2022 at 6:13 | vote | accept | Anaconda | ||
| Feb 17, 2022 at 18:15 | history | edited | Richard Hardy |
edited tags
|
|
| Feb 17, 2022 at 18:12 | comment | added | Richard Hardy | I have posted answers to perhaps half a dozen related threads. One of them is this. Some more are here; you may wish to narrow the search down by adding additional keywords. | |
| Feb 17, 2022 at 18:08 | answer | added | Richard Hardy | timeline score: 1 | |
| Feb 17, 2022 at 2:37 | comment | added | Anaconda | 𝑥1 which is I(1) according to the ADF test, 𝑥2 which is I(2) according to the ADF test, x3 which is I(1) according to the ADF test, In order to let all three to be stationary, I finally conduct all Xs , which are I(2) according to the ADF test. Then determine p from VAR and use it to build VECM model. | |
| Feb 16, 2022 at 9:25 | comment | added | Richard Hardy | Could you please write this down in detail? E.g. say: I have $x_1$ which is I(2) according to the ADF test or whatever other way you used to determined the order of integration; $x_2$ which is I(1) according to ...; $x_3$ which is I(0) according to ...; and so on. Just to make it absolutely clear what you have. Then it will be easier to take it from there. | |
| Feb 16, 2022 at 8:28 | comment | added | Anaconda | Some of them I(1)，some have to I(2) to make them stationary. | |
| Feb 16, 2022 at 6:59 | comment | added | Richard Hardy | Are all of your time series I(2), or are some of them I(1) or I(0)? | |
| Feb 16, 2022 at 2:20 | history | edited | Anaconda | CC BY-SA 4.0 |
added 595 characters in body
|
| Feb 16, 2022 at 2:12 | comment | added | Anaconda | The transformations of the variables is normalized by MinMax, whose range is between 0 and 1. | |
| Feb 16, 2022 at 2:08 | comment | added | Anaconda | integrated of order two (Δ2𝑥𝑡) for all series. My data is not only integrated of order one, but must two to keep all series stationary. According to Lutkepohl "New Introduction to Multiple Time Series Analysis", it only gives proof of integrated of order one, but no proof more than one. When I get through the chapter, it seems that Δd Xt works. Because so far I am not writing paper but just the aim to conduct a stable model without residual autocorrelation and not-bad generalized model. To some degree, the model need not be so strict with theory or mathematic proof. | |
| Feb 15, 2022 at 20:03 | comment | added | Richard Hardy | Could you describe what you are doing in a bit more detail, e.g. using symbols such as $\Delta^2 x_t$ for the second difference of $x_t$? What the variables are, what their orders of integration are, what transformations of the variables are included directly into the model. | |
| Feb 15, 2022 at 12:59 | history | asked | Anaconda | CC BY-SA 4.0 |