VECM. Model Two. Part 3 of 4. VECM

[@sayedhossain23]

Trace and Max test both are equally valuable, so you can take result of any one by taking it as benchmark. You can not run both VAR and VECM. Either VAR or VECM.

[@samermehibel1324]

2- A s i am interested in studying the impact of fiscal policy on macroeconomic variables, normally i use the IRF to study that. So my question is should i ignore the trace test and ignore the fact that there is 2 cointegrating relationships and adopt the results of Max eignevalues test and act accordingly and estimate a VAR model so i will be able to study the IRF? or if i adopt Trace test meaning 2 cointegrating relationships, have i the right to study the IRF from the VECM model?

[@erasmushyera3910]

Thanks sir for feedback , i have run johansen test cointgration and find 1 coint relation, should i use VAR or VECM? , i want to obtain beta for real gdp per capita and m2/gdp and real gdp percapita and credit/gdp in separate VAR. should i take the normalised value and the beta from the target dependent variable real gdp percapita? and what if are longrun coefficient is negative but not signficant should we reject model? or is it an implication of shortrun cauality or weak causality.

[@ferhanarslan4015]

Hello, thank you for your work. How to do and interpret the Frequency Granger Causality test with Eviews program.

[@palwasha3803]

Part 4 of 4????

[@lexchuanlixia815]

Hi, Dear Dr. Hossain, thanks for your video and I find it really helpful. Generally, I have a question regarding the long run effects. In this video, you demonstrate that there is a long run causality running from CONS and EX to GDP, but I am wondering what the exact coefficients of such long run effects. In the first part of the D(GDP) equation, there exists an cointergating equation, C(1)*(GDP(-1)-0.16*CONS(-1)+0.54*EX(-1)-47.78). Therefore, I wanna know the coefficients of long run effects of each IV for example, the CONS. So is 0.16 the coefficient of long run effect of CONS on GDP? Or should I multiply C(1) with 0.16 to get the coefficient? Also another question is that since the coefficient of GDP(-1) is 1 in the cointegration equation, does this mean that the long run effect of GDP on D(GDP) is 1? Or actually there is no such long run effect of a variable on its own first difference? In this example, there exist no long run effect of GDP on D(GDP). Thanks so much for your attention in advance and look forward to your reply!

[@sayedhossain23]

Thank you. I would like to invite you to join Hossain Academy Facebook at below link and post your question there for feedback. Thank you, Sayed Hossain from Hossain Academy facebook.com/groups/hossainacademy/

[@sayedhossain23]

As you have got 1 coint equation, you should run VECM, not VAR. Regarding beta I have no idea

[@palwasha3803]

If we have 2 con integration equation?

[@palwasha3803]

Then what we do plz tell me sir

[@HB-gn8kr]

I have a question, in case we have 2 cointegrating vectors, we will have 2 coefficients representing long-run causality, if one was significant and negative while the other was only negative but not significant, how can we interpret that result? thank you

[@sayedhossain23]

Hasan Baalbaki In that case it will be hard to decide. I would suggest you try to get one cointegration vector, not two.

[@HB-gn8kr]

Sayed Hossain This will be possible specially that when doing trace test I found 1 cointegrating equ for a significance level of 1% and 2 for a significance level of 5% , with max eigenvalue test giving me 1 cointegrating equation for 1% and nothing for 5%. In that case can I carry on with one cointegrating equation? Thank you, appreciated

[@abujumana7490]

Hello dear If Granger causality refer to no causality relationship, but the VECM refer to one long run relationship, can I adopt the both models in my thesis? and explain the both results regards

[@sayedhossain23]

You need to use one model. Two model at a time will bring conflicting result.

[@abujumana7490]

OK Dear, thank U very much

[@sayedhossain23]

In a thesis I would suggest to take one model to discuss, not two.

[@sayedhossain23]

I said already...Now u decide..

[@abujumana7490]

Sayed Hossain Thank U dear, my thesis is about the relationship between exchange rate and oil price, I used monthly data series, length lag criterion refers to (2 lag), what I did, I ran the causality test for four first lags (i.e, 1, 2, 3, 4) in the first 3 lags there was no causality relationship, but in the fourth lag there was one direct relationship from oil price to exchange rate, I documented the all 4 state in the thesis and also I used the VECM which refers to the same direct of relationship (i.e. from oil price to exchange rate in the long run) I want to ask if this is OK? Thank U very much, I adopted your videos in my empirical framework. and I mentioned U in the thank page of my thesis. Best Regards

[@yesuusfayyisaadha231]

Dear family, How are u? In my Analysis i have reached a VECM and i did wald test to check whether variables coefficients are zero or not, but How can i find significance of the long-term variables significance? And should i take the opposite of the coefficients sign while building my long-run equation? Thank you in advance! Actually my speed of adjustment is negative and also significant, but what i want is individual variables significance, and my additional question is how i do normalization of long-run variables. How can i find normalized α and β ?

[@sayedhossain23]

Would you mind to join Hossain Academy Facebook below link and post your questions. Normally we are sharing there our discussion.

[@sayedhossain23]

facebook.com/groups/hossainacademy