Econometrics # 38 : Error Correction Model with EViews

[@TJAcademyofficial]

Negative relation between change in Y and Lagged error term show long run adjustment.

[@geetanjali3436]

Your work motivate me Thanks a lot

[@callankellan4955]

I dont mean to be so offtopic but does any of you know a way to log back into an Instagram account? I was dumb forgot the password. I would love any tips you can offer me.

[@callankellan4955]

@Edward Luca i really appreciate your reply. I found the site on google and im trying it out now. Takes quite some time so I will reply here later with my results.

[@callankellan4955]

@Edward Luca it worked and I actually got access to my account again. Im so happy! Thanks so much you saved my ass!

[@edwardluca6713]

@Callan Kellan glad I could help :)

[@tondeutsi4195]

10/10 you've managed to turn a complex and slightly boring area of economics into something easier to understand and interesting. you're not the hero we asked for......but you are the hero we all needed

[@TJAcademyofficial]

Thank you for your appreciation ☺️

[@takmilahsan]

Thanks for making Econometrics so easy for students. Typical subjects become interesting & easier just because of teachers. جزاک الله

[@SkateboardingPakistan]

Can be apply the same procedure to panel data?

[@weena1725]

Question: The VEC model have two cointegrating equations. The coefficient, C(1) is negative and significant, but C(2) is positive and significant, what will be the overall conclusion for longrun equilibrium?

[@thabitiyssoufa3232]

Thanks for sharing ,you made it easy .

[@kalpana2545]

Sir can you please provide a separate video on ECM without Eviews ,Only explain the theoretical portion ASAP

[@chandnirana369]

You are one of my favorite sir even of my friends 😇😇😇😇😇

[@samfisher1250]

i have a and b variables and i used johansen cointegration and it shows that there is 1 cointegrating vector. when i tried using engle-granger approach to it using a as a the dependent variable, the ect is insignificant. however when i exchanged and used b variable as the dependent, it shows significant and negative ect. what does it mean?

[@NilofarAsgari]

If I run regression with log and the error is not stiosinary what i can do?

[@geetanjali3436]

Sir you are doing greate job If your video is not available I can't even think to learn econometrics

[@IrfanullahSahibzada-v4d]

Sir I am confused about when ECM model got estimated then how to forecast for the specific dependent variable while ECM independent term residuals are not available for the forecast period.

[@rioprasetya5280]

Sir, if short term all variable show not signifikan, can we lag all variabel too ?

[@240419921000]

Sir if gamma is positive n insignificant, what is the interpretation?

[@KailaJaneBermudo]

What happens when one of the variables is stationary at level? Can I remove the d in equation? Or can I proceed to ecm?

[@farwahali2930]

Well explained Sir thanks.

[@jhabindrapokharel]

A simple way of explanation of the matter impressed me. Thank You, Dr Sab. Could you please make a video on the structural VAR model?

[@michaelasare4987]

@TJAcademyofficial, I learnt we can't use the test statistics from the ADF on the Unit root test for the error term.

[@TJAcademyofficial]

Reference plz. I want to read what exactly it is.

[@sanaamami]

if i have a VECM lag 4 how can I calculate the short and long run

[@aneezamuhammadzafar4724]

Expressionless No words to express thanks

[@shanmohammad4193]

well explained Sir, please keep it continue and covers all latest model i.e ardl , and nonardl

[@TJAcademyofficial]

JazakAllah 🙂

[@fbatravels]

Nice effort

[@ayeshasabir2373]

When conducting the ADF test for residuals, some options exist for "include in test equation" : intercept, trend & intercept, & none. Why did you leave intercept term selected? Shouldn't we select the option none? How do we know what option to select because in some cases the test results change from a rejection to cannot reject just because of these options.

[@TJAcademyofficial]

Thank you for your message. You can take all three options for robust result.

[@abiwugodson7651]

Thanks, very helpful

[@ayadhichem4567]

really great video thanks a lot

[@TJAcademyofficial]

My pleasure

[@anupamsabharwal4385]

Very well explained thanks a lot

[@aminaahmedalibelal5676]

Do you mean if variables are I(0) and ECT is I(0) means there isn't a cointegration (long term relationship)?

[@TJAcademyofficial]

If all variables are I(0) then there is NO need to check cointegration

[@nadeemacademy5262]

Awesome sir MashAllah

[@geetanjali3436]

Sir I have run VECM residuals diagnostic but my model found non normal and hetroskedastic residuals but it solution for it I already taking my variable as natural log form. What can I do for this problems Pls rpy

[@okanaybar]

wonderful teaching. thanks from turkey

[@TJAcademyofficial]

My pleasure 😊

[@ashveenaashveena570]

Sir what if we have already differenced the variables, do we still have to write d(variable)?

[@TJAcademyofficial]

No

[@ashveenaashveena570]

@@TJAcademyofficial Thanks alot for the reply, I just have one more question. If I apply VECM or ARDL then do we have to check autocorrelation too? and if it exists then, should we show the result we got after removal of auto correlation in our research?

[@maharnadeem-ix1hd]

Sir I really impressed your work

[@fahadsultan8473]

Great sir g , kindly in models ki forecasting pe b ik video plz

[@NirmalaBhatt-xm9op]

Kindly requested to make video on #Gregory-Hansen Cointegration Test and VECM.

[@calebchipeta3781]

I still would like to understand how you would interpret a speed of adjustment between -1 and -2 .like you were able to say 0.3 means 30 % what of -1.7

[@TJAcademyofficial]

I have discussed it in last few minutes of the video

[@calebchipeta3781]

@@TJAcademyofficial respectfully doctor you speak about the Error correction term at 14.9 minutes of your video and at 17.8 minutes. In the first case you talk about the coefficient of the error correction term and you say if it is from o to 1 in absolute value , then it goes to equilibrium directly, but if it is -1 to -2 it oscillates to equilibrium. in the second case you explain how to interpret the coeffrient of an error correction term. you found -0.27, and rightly said the speed of adjustment is 27 % per unit time. My quetion is how do you interpret(NOT THEORETICALLY), an error term of -1.7, do you say adjuustment hapens at 170% unit time , because thats confusing , plrease elaborate sir

[@TJAcademyofficial]

Yes it is the case of 170% which shows that after one time adjustment values have crossed the equilibrium value and in second time it will move towards equilibrium again. After some oscillation equilibrium will be achieved.

[@pradeepdangi9614]

Thank you sir, for interpretation of ECT as lies between -1 and -2. I have estimated ARDL model and in my short run dynamics estimation, ECT was -1.3633 and significant at 1% level but my thesis supervisor said me "ECT have gone more than 1(with negative sign) so this model may not be reliable." Therefore, i was searching the interpretation when ECT is between -1 and -2 . I saw this video and found really helpful to me for interpretation. Will you send me any article or book name that has interpretation of ECT with value falls between -1 and -2, so that i can present to my thesis guide as an evidence?

[@TJAcademyofficial]

Share your WhatsApp

[@TJAcademyofficial]

You can see my answer with reference below: www.researchgate.net/post/Problem-with-Error-correction-coefficient

[@michaelasare4987]

Why ECM uses the lag of the Error term?

[@TJAcademyofficial]

All explained at the last part of the video.

[@michaelasare4987]

@@TJAcademyofficial Yes, I understood from you the video why the error term is used in the model. But why didn't we use error at t but t-1?

[@aminaahmedalibelal5676]

In step one, what if variables are also stationary ?

[@rajasingh4144]

Sir please you have eview setup please share

[@usmansaleem1253]

sir please VECM pa b asi hi comprehensive video bana dain secondly differnce of ECM and VECM wali b

[@abhishekkacholia3864]

Sir If the data is stationary at first difference but there is no cointegration between the variable i.e there is no long run relation then which method should be used for testing short and long run relationship between the variables? Restricted VAR or Unrestricted VAR? Can I use VECM model?

[@TJAcademyofficial]

VECM used when cointegration exist

[@abhishekkacholia3864]

Sir if there is no cointegration then which model should be used?

[@TJAcademyofficial]

OLS with stationary condition or VAR

[@abhishekkacholia3864]

Okay Sir Thank you for helping 🙏

[@mbkhan8418]

AOA sir. U are my favorite teacher. Can u please make a videos in future covering on these topics as under. 1. Baumol's Theory of Sales Revenue Maximization 2. Marris Hypothesis of Maximization Growth rate 3. Williamson Model of Managerial Utility Function 4. Principal Agent Problem (asymmetric Information)

[@TJAcademyofficial]

Ws. Thank you for your message. I am currently working on econometrics. Will try my best to cover these topics asap.

[@nsakib62]

Sir is it necessary to cointregetd all the variable before ECM test

[@TJAcademyofficial]

Ye It is. ECM shows adjustment towards long run. If long run or cointegration does not exist then adjustment towards equilibrium has no meaning.

[@tariqrahim1129]

Dear Sir, Can you please estimate ARDL-ECM in EViews?

[@incikara3210]

Hello sir, I hope you will answer my question. I found error correction form negative sign (-1.07) and statistical significant. Is there any problem with that? Because it’s not in the range of 0 and -1.

[@TJAcademyofficial]

Hi, watch between 14:35 to 16:05 of this video. It can be -1.07 which shows long run adjustment will be in dampening manner

[@incikara3210]

@@TJAcademyofficial Thank you for the answer. I’m just confused and fade up. There are so many ideas about this topic I have a thesis to finish and I found that result so I don’t know how to explain it well. Could you recommend an article which explain ect for the range of -1 and -2? Thank you so much sir.

[@tehseenjawaid8393]

@@incikara3210 What Determines Migration Flows from Low-Income to High- Income Countries? An Empirical Investigation of Fiji-U.S. Migration 1972-2001

[@etc4363]

Highly beneficial content. Alright, this way we can find out short-run impact and relationship and adjustment towards long run as well. Although ECT shows adjustment towards long-run relation and error term significance shows the existence of the long-run relationship. But in this context how do find out the long-run impact?

[@hermainsarfaraz3167]

A little different from this topic. But could you tell that if we estimate an Arch model(ML approach), and it's significant, what will it show?

[@madhvikaushik5297]

Sir plz make a video on How to download quarterly data of GDP from world bank website Thank you

[@abiwugodson7651]

Please in all of these, where can I place the control variables?

[@TJAcademyofficial]

Thank you for your message. You can add control variable as a independent variable.

[@TJAcademyofficial]

Thank you for your message. You can add control variable as a independent variable.

[@nomanrasheed9644]

sir can you provide the reference of ECT term between 1 and 2.

[@pawankumarsingh2114]

nice

[@rafeequeahmed7282]

sir, how to explain significant error correction coefficient ranging between -1 to -2, kindly help?

[@TJAcademyofficial]

Moving around equilibrium value in dampening manner as discussed in last of the video

[@rafeequeahmed7282]

@@TJAcademyofficial thank you, sir.

[@vitaltopics316]

@@TJAcademyofficial 1) what we need to do if the ECT has I(1)? what is the next thing to do? 2)can we use the model for panel data N=11 and T=12, or N>T, or N

[@shabbarimam4779]

Assalam o Alikum Sir! if during Bound test the probability value of variables is greater than 0.05 so in this case bound test results consider wrong?

[@sanasehar4878]

Plz kindly make a video on ARDL model e views application

[@zulfaqarkhan9285]

Sir you are great ,thanks a lot ...sir plz also make video on these topics plzz Cobb Douglas production function Perfect complement Leontif technology Perfect substitute linear technology Constant Elasticity of substitution

[@zulfaqarkhan9285]

Thanks a lot sir ....sir aik lecture Matrix algebra pr bnao ...Econometrics mai Y= beta X+ e

[@TJAcademyofficial]

My pleasure. Please find the links below on OLS and Econometrics Model ECONOMETRIC Model: kzbin.info/www/bejne/jZmUknqGjLuCqNE OLS: kzbin.info/www/bejne/rpXQqn2LbbiXhs0

[@zak26khan]

what is the difference between ECM and VECM?

[@TJAcademyofficial]

ECM is for single equation VECM for multiple equation

[@mukaramazhar3847]

VECM model. Which equation should be included in research paper. also help in conclusiin writin which equation should be used as conclusive like showing the relationship among variables from where we conclude the overall results

[@usmansaleem1253]

sir please make me clear that title of lecture is ECM in eviews. can we also call it VECM? sir its a confusion please clear it

[@TJAcademyofficial]

No

[@usmansaleem1253]

@@TJAcademyofficial sir vecm pa b lecture dain k wo kab use kartay hain

[@usmansaleem1253]

@@TJAcademyofficial sir as per my knowledge Johanson k sath Vecm lagtti ha or ARDL k sath Ecm

[@zoyashah7826]

Sir if I am applying directly ARDL model to estimate the short run dynamics and not using least square method, I am getting 3 coefficient values of some variables as I have taken the lag value 3..then which one to consider to mention in my research paper??

[@ashveenaashveena570]

all of them and you can also try reducing to lag one

[@zoyashah7826]

@@ashveenaashveena570 how I can reduce it to lag 1??

[@ashveenaashveena570]

@@zoyashah7826 when you apply ARDL there comes a window which asks you about it, just write "1" there

[@ashveenaashveena570]

@@zoyashah7826 just so you know, whenever I used 2 or 3 lags, I interpreted only about 1 lag .

[@aemonenam2205]

What if the data of few variables are stationary at 2nd level? Then which model will be used to find out short run please reply me

[@ashveenaashveena570]

toda and yamamoto

[@HishamMahran-r9k]

Data set please