Dear researchers,
I am interested in studying the long-run relationship between the panel datasets. However, most of the empirical works used the PMG/ARDL technique to test the long-run parameters of the variables and the error correction term of the model. But they failed to confirm the long-run relationship by using appropriate panel cointegration techniques (Eg. Padroni/ Kao test). In the meantime, some studies employed the panel cointegration test before applying the PMG/ARDL technique. They argue that the PMG/ARDL technique is only used to resolve the heterogeneous problem. Thus, I want to know that panel cointegration is needed to apply before PMG/ARDL technique.
Although it is not necessary, but if you do that it will make your model more robust
If variables are stationary, it is cointegrated. To test long run relationships, still imply variables are cointegrated! In PMG/ARDL, testing stationarity is important before running ARDL/PMG.
Wish you good luck
Dear Ahamed.
For a robust result, you should carry out a cointegration test to verify if the variables are cointegrated before moving on to PMG/ARDL techniques.
Regards
Of course, it is needed since the tests are purposely employed to show the extent of sensitivity of two variables to the same average value over a period of time. I don't know why someone will say it is not necessary when this is important for the avoidance of bogus results and making the model to be more wholesome. In the current research world, everything deem necessary should be done in order to present reliable and not deceptive results.
I believe the panel bound test of cointegration should be enough to confirm long run relationship in your model if indeed there is. So try that.
Bekoe Bernard Boamah, where/how can I find a panel bound test for Stata, and/or Eviews? Or R?
- Badges
- Science topic
- Similar topics
- Linguistics
- Composition Studies
- Publication