HOW TO DO ARDL MODEL PART I EVIEWS

[@abidemisomoye]

You explained it so simple. Thank you so much Dr.

[@nasir467gcu]

I love ur videos

[@nasir467gcu]

Madam i need hlp in analysis can yiu spare some time

[@angelaadomokhai4388]

Thank you for this. Please after running my ARDL ECM, I have only the constant and the error correction term, and no short run differenced variable. What does this mean please? Also can i still go ahead and explain the long run relationship since my f-statistics is greater than my I(1) bounds

[@dr.shobhak6764]

Post your results via [e-mail]

[@tosin_davidson]

HI Dr. Im using Eviews 9 and I used ARDL to get a long run with cointegration rresult. There is no ECM in the ARDL drop box.,. Can I use OLS in the estimate equation to run a short run ECM test separately

[@showrhov]

which version you use? my version is not similar to you.

[@donasp5391]

Thank you! Could I use ARDL if I have a small T (10 years) and a large sample (360 companies)?

[@dr.shobhak6764]

Just check the stationarity and then decide on ARDL

[@donasp5391]

@@dr.shobhak6764 Thank you so much!

[@thewaldwiesel]

When do I have to use 1.) none 2.) Rest. constant 3.) constant 4.) Rest. trend or 5.) Const. + trend. in the settings for the trend specification?

[@dr.shobhak6764]

Pls refer the eviews manual. Since I will be giving the same answer

[@iwasamegbe6177]

when performing unit root test, do i have to log the variables or i just test for unit root with the main variables

[@dr.shobhak6764]

Usually log is taken

[@seeratsajjad3224]

Is multicollinearity annissue incase of Ardl model or can it be ignored ?

[@dr.shobhak6764]

It cannot be ignored

[@LifelongStudentBelgium]

I don't really understand, you say your model shows short run relationships? this is partly correct, it is so that the coefficients are almost identical to the EC-model that you can interprete after bounds testing has been found. But what if the EC-model is not showing a variable that was significant in the original ARDL?

[@dr.shobhak6764]

My dear student you are very much confused

[@dhakaramkadel3671]

Good afternoon, madam. Would you please see this also? While conducting the ARDL model with four variables in Eviews, all the variables are non-stationary at level, but at the first difference, i.e., I(1), two variables, including the dependent variable, are stationary without taking their log, but two other independent variables are with log only. Should I need to use a log for all the variables in this situation?

[@dr.shobhak6764]

Use log for all the variables

[@dhakaramkadel3671]

@@dr.shobhak6764 Thank you very much mam

[@najeebkhan4246]

what if my dependent variable values consist of positive and negative values which model should i use? all variables are stationary at 1st difference. can you help me.

[@SherMehta-l4p]

If all the three variables are 1(1), one should first run a cointegration test and only if they are cointegrated, should the above variables be in level form. Otherwise, the ARDL model should be in the first differences. Please explain why the ARDL model is being explained in levels without a cointegration test. If the cointegration tests fails, then the variables should be in the first difference and then they should be interpreted. Kindly explain.

[@Uttam105211]

Madam, why have you used the first difference of the variables?

[@tosin_davidson]

Hi.. thank u.....

[@tosin_davidson]

HI Dr. Im using Eviews 9 and I used ARDL to get a long run with cointegration rresult. There is no ECM in the ARDL drop box.,. Can I use OLS to run a short run ECM test

[@varshneyvikas39]

Can we take series at first difference? Please clarify

[@dr.shobhak6764]

Yes

[@varshneyvikas39]

@@dr.shobhak6764 if we take stationary series then what is the purpose of checking cointegration?

[@dr.shobhak6764]

@@varshneyvikas39 If the variables are stationary then it can't be cointegrated

[@belabela4011]

Which model we can use instead of ARDL model???

[@dr.shobhak6764]

Under which circumstances?

[@belabela4011]

time series econometrics.. There are 5 variable.. Which are stationary at I(0) and I(1)..What is the disadvantage of using VAR model??

[@zoyashah7826]

Mam please tell that if m using RBi database,then there is data for different base year.so how I can calculate the data for the different base years ??

[@huyau2003]

Hi when i convert variables into first different I can't run ARDL because it said log of non positive number, can you help me solve this problem?

[@dr.shobhak6764]

Can't take log for negative number

[@huyau2003]

@@dr.shobhak6764 the normal log variables are not negative but the first different log variables are negative? I watch your tutorial to convert variables into first different and the variables in those are negative too, im confused.

[@belabela4011]

Mam can you please help me for my data to set the ARDL??

[@dr.shobhak6764]

I have got a hectic schedule. Kindly send it via email. But don't expect me to give it on time

[@belabela4011]

@@dr.shobhak6764 tnx mam..please give me the email..