I have a mix order of integration after doing the unit root tests for all my variables. That is I(0) and I(1), I can't run panel Johansen co-integration or panel least square fix/ random effect because of this. it would violate the condition or assumption underlying them. The suitable approach is panel ARDL using Eviews-11. But i can't find any diagnostic test except for Histogram normality test. I don't know how to carry out serial correlation LM test, heteroscedasticity using the panel PMG/ARDL method on Eviews-11. Can i run the diagnostic test using the ordinary regression? and still use ARDL? PLEASE HELP.
Perhaps staring here may be of some use? https://www.google.com/search?client=firefox-b-1-d&q=Autoregressive+model+history+Yule
Best, D. Booth
PS you might also try: https://www.google.com/search?client=firefox-b-1-d&ei=l42pX7jqDNSStAbcnoTQBA&q=Eviews+and+ARDL&oq=Eviews+and+ARDL&gs_lcp=CgZwc3ktYWIQAzIJCAAQyQMQFhAeMgYIABAWEB4yBggAEBYQHjIGCAAQFhAeOgQIABBHOgcIABDJAxBDOgQIABBDOgIIADoFCAAQyQNQl74FWNHjBWDS6AVoAHACeACAAXuIAbIHkgEDMS44mAEAoAEBqgEHZ3dzLXdpesgBCMABAQ&sclient=psy-ab&ved=0ahUKEwj4kYLmjvbsAhVUCc0KHVwPAUoQ4dUDCAw&uact=5
Hello my sister Kehinde Hammed , with mixed order of integration I(0) and I(1), the additional pre-estimation diagnostic test you need is the ARDL Form and bounds test of cointegration. If your variables do not have cointegrating equation, you will go ahead to conduct ARDL, but if you have long-run cointegration, then, you will be required to conduct Error correction ARDL which is a short-run and long-run ARDL The post estimation/diagnostic test you you can conduct with the Eviews are Jarque-Bera Normality test serial correlation LM test, heteroscedasticity etc.
However, if your data is a panel you can use the panel Pool Mean Group ARDL I have not seen any post estimation test on PMG ARDL in Eviews. I think the reason is because of the assumption that method restrict serial correlation.
Godwin, why is it assumed that PMG ARDL in Eviews restricts serial correlation?
Dear Kehinde, since you are making use of PMG/panel ARDL, check for cross-sectional dependence. If there is no cross-sectional dependence, proceed to check for cointegration using the pedroni. Kao, etc. and if there is cointegration, go on to interpret your results.
However, if there is cross-sectional dependence you will need to utilize the second generation test of unit root, the westerlun cointegration test which are available in STATA.
If you run Traditional Panel, then the following diagnostic tests have to run: Hausman Test, Normality Test, Multicollinearity Test, Hitroskedasticity Test, Serial correlation. Linearity Test. If you run panel ARDL, then in addition to the above tests, you have to run Unit root test.
https://blog.eviews.com/2017/05/autoregressive-distributed-lag-ardl.html
- Badges
- Science topic