Dear colleagues,
Currently I'm doing research using ARDL-ECM method processed through Eviews 10. Can somebody mention or explain what are the robustness test used for ARDL-ECM analysis?
Thank you
Dear Firsty Izzata Bella , you can apply the DOLS or FMOLS models to test the robustness of your results.
A little more clarity is needed. However, the post estimation test required for ARDL ECM are Test of; Heteroschedasticity, Serial correlation, normality, and stability, etc
Dear Firsty Izzata Bella
To perform the robustness test, you must make changes to the regression and compare the results of the new estimate with the results of the initial estimate to see if the initial results are valid.
This test is to check the validity of your estimation results
You make changes to regression or model or estimator or variables and re-estimate after making changes
For example, change the estimator, or add a control variable, or change your explanatory variable
If the results of the new estimate, except in the size of the coefficients, confirm the initial results (for example, the sign of the coefficients or its asymmetry), then your initial results cannot be rejected.
If you still have problems, I will explain more
Dear Firsy,
You can consider the following for robustness test.
1. Serial correlation
2. Heteroscedasticity
3. Ramsey reser
4. Normality
Then you can also USE below to confirm the stability of your parameter
5. CUSUM test
6. CUSUMQ test
Regards
Except for normality, I am in agreement with Kehinde Mary Bello. Please I am not arguing that normality isn't important but it is minimal according to monte Carlo studies.
- Badges
- Science topic
- Similar topics
- Biological Science
- Histology
- Sectioning