VECM. Model Two. Part 3 of 3. STATA

[@Fredekkkkkk]

First difference since all my variables are integrated order 1. Thank you very much for your help

[@asmamawgetnet5343]

Hello professor I thank you your clear lecture but I have one question regarding with vec diagnostic test. veclmar is not working

[@sayedhossain23]

The standard guideline is, make the data stationary first (in that case u can covert the data into first differenced to make it stationary) and then run regression.

[@Fredekkkkkk]

I am trying to explain the changes in log of prices of gold using selected a number of independent variables (lnS&P500, lnVIX, lnOil_Price, lnUS_Inflation and lnDollar_Index) I have followed your steps for testing for long run casualty & short run casualty on my model. Using VECM I found the Error Correcting Term to be positive, am I correct in understanding that this means there is no long run equilibrium? Where can I go from here with my model? Can I run OLS on first differences?

[@sayedhossain23]

You are welcome

[@Fredekkkkkk]

Thank you! :) Do you think its beneficial to run OLS on first difference?

[@sayedhossain23]

I did not attempt to explore it yet. So can not answer now

[@sayedhossain23]

It should be negative to have long run causality. You may not have long run causality but I guess you have short run causality from there.

[@minyahilalemu4302]

Great video, it helped me as well. Thank you for that. Just one question: in case where there are more than one co-integrating equations, the long run result or the Johannes's normalized equations in the VECM omits the variables in sequential equations. Please, what shall?

[@sayedhossain23]

+Minyahil Alemu Yes it may have more than 1CE in Johansen test but when you run VECM model, use 1CE

[@minyahilalemu4302]

+Sayed Hossain Hi, sir Thank you for your help. But doesn't it matter to use only one cointegrating equation in the existence of more than one?

[@sayedhossain23]

+Minyahil Alemu If more than 1CE there is no way to interpret results..But with 1CE we can interpret results nicely..

[@minyahilalemu4302]

+Sayed Hossain thank you, once again. got much!

[@sayedhossain23]

+Minyahil Alemu Dear Alemu,, Thanks for appreciation. I would suggest you to join Hossain Academy Facebook at below link and post your question there. If I know the answer I shall respond. Thank you once again, Sayed Hossain from Hossain Academy facebook.com/groups/hossainacademy/

[@pianogirl0617]

Hi Prof, I found your video are very useful to understand the numerous coefficients from the VECM. Just one question which bugged me for a long time. How to interpret the sign appeared in the tables? I heard that to put them into equation there needs a "-" in each of the coefficient? Is that valid? Thank you so very much.

[@universalgain999]

What is the name of the post estimation test you have used? The Chi-square score was used in place of an F critical value

[@tadasrirathung9492]

Professor, Please give the video on how to use VECM results to do in sample forecast. Thanks

[@sayedhossain23]

Dear Tada, Thank you. I would like to invite you to join Hossain Academy Facebook at below link and post your question there. Actually I am in that group and may help you. Thank you once again, Sayed Hossain from Hossain Academy. facebook.com/groups/hossainacademy/

[@AlvarezMoon0710]

Hi Sayed Hossain, I have 1 questions: What is the meaning of "johansen normalization restriction imposed" table? In your video, I don't see you have any mention about that table. Because I'm newbie of STATA, so sorry if my question is simple.

[@metal89dc]

Hi sir, I have a question a question regarding your video Should the error correction term be negative and between -1 and 0? Thank you

[@sayedhossain23]

+Andrija Vesic Yes you are right. However, Thank you. I would like to invite you to join Hossain Academy Facebook for greater interaction about economics, finance and econometrics with me. Thank you Sayed Hossain from Hossain Academy. Please join below and post your question.facebook.com/groups/hossainacademy/