Hello everyone,
My question will sound easy but I get confused by reading so much different things :
When you have datas and you don't know if its stationary or not, apply dfgls test on data, and you get something like this :
DF-GLS for x Number of obs = 29
Maxlag = 9 chosen by Schwert criterion
DF-GLS tau 1% Critical 5% Critical 10% Critical
[lags] Test Statistic Value Value Value
------------------------------------------------------------------------------
9 -0.827 -3.770 -2.719 -2.372
8 -0.985 -3.770 -2.757 -2.428
7 -1.216 -3.770 -2.818 -2.502
6 -0.829 -3.770 -2.896 -2.586
5 -1.100 -3.770 -2.986 -2.678
4 -2.275 -3.770 -3.081 -2.771
3 -3.021 -3.770 -3.175 -2.861
2 -3.056 -3.770 -3.262 -2.943
1 -2.928 -3.770 -3.336 -3.011
Opt Lag (Ng-Perron seq t) = 5 with RMSE .6426388
Min SC = -.3943049 at lag 1 with RMSE .7310553
Min MAIC = -.2705526 at lag 6 with RMSE .6215392
How can it helps me concretely to know how to make my data stationary?
I usually specify directly ARIMA model on data, then test "estat aroots" and then I know if my process is stationary or not, but I think it is not the same because here I test if the process I specified is stationary, but not the data .
If someone can explain me this it will be so helpful
Thanks in advance,
Jean
My question will sound easy but I get confused by reading so much different things :
When you have datas and you don't know if its stationary or not, apply dfgls test on data, and you get something like this :
DF-GLS for x Number of obs = 29
Maxlag = 9 chosen by Schwert criterion
DF-GLS tau 1% Critical 5% Critical 10% Critical
[lags] Test Statistic Value Value Value
------------------------------------------------------------------------------
9 -0.827 -3.770 -2.719 -2.372
8 -0.985 -3.770 -2.757 -2.428
7 -1.216 -3.770 -2.818 -2.502
6 -0.829 -3.770 -2.896 -2.586
5 -1.100 -3.770 -2.986 -2.678
4 -2.275 -3.770 -3.081 -2.771
3 -3.021 -3.770 -3.175 -2.861
2 -3.056 -3.770 -3.262 -2.943
1 -2.928 -3.770 -3.336 -3.011
Opt Lag (Ng-Perron seq t) = 5 with RMSE .6426388
Min SC = -.3943049 at lag 1 with RMSE .7310553
Min MAIC = -.2705526 at lag 6 with RMSE .6215392
How can it helps me concretely to know how to make my data stationary?
I usually specify directly ARIMA model on data, then test "estat aroots" and then I know if my process is stationary or not, but I think it is not the same because here I test if the process I specified is stationary, but not the data .
If someone can explain me this it will be so helpful
Thanks in advance,
Jean
