Engle-Granger approach to cointegration.avi

[@murat88yas]

1)Open variables as group. 2)View->Cointegration Test->Single Equation Cointegration Test 3)Choose Engle - Grager Done

[@nicolettelujaherrera1548]

Yes, but how do you interpret the results?

[@lucnguyenthanh4414]

I think for ADF test you should choose none rather than intercept or constant because the residual is estimated rather than observed and the most important assumption of OLS is that the mean of the residual should be equal to zero (None)

[@superthiru]

Thanks so much!! Coming from a finance/accounting background, I was clueless on how to do this, but you explained it in very simple terms. You're a legend!

[@leoneltclemente]

To perform the Engle-Granger test, open an estimated equation and select View/Cointegration and select Engle-Granger in the Test Method combo. The dialog will change to display the options for this specifying the number of augmenting lags in the ADF regression. (eviews guide II, p.238)

[@nicolettelujaherrera1548]

Hi, do you know how to intepret the p values' results?

[@Seikhupatz]

Very helpful video! Pros: succinct, beginner friendly, clear voice, good coverage of material and meaning of analyses Cons: video quality could be a bit better but not really necessary.

[@nwachukwunnamdi6174]

According to Gujarati, You have to run the first model including the trend, then you extract the residuals from there. you can check this out from Gujarati Econometrics text-book.

[@adrenalinerush369]

Thank you for the tutorials. It really helps to have it shown step by step. If it's not too much of a bother, I would like to respectfully request for the links of the materials (ie. tables of critical values) be included in the video description for faster following of the student. Other than hopefully a little better visual clarity (my eyesight is very poor), I would say a very nicely made video that is easy to follow. Many thanks!

[@sanjuzameena3176]

there is no cointergration mean that there is no long run relation so what happen next? what are the further steps?

[@michaelpatterson8219]

Thanks for your explanation of the Engle -Granger test.

[@seung-hyunoh6616]

Thank you very much, this help me a lots in my bachelor degree final year research.

[@shaloms9640]

Which variables should I use for Johansen's test and estimating the Vector Error Correction Model? level or the differenced values? If my dependent variable is exchange rate (ER) for instance and my dependent variables is price of copper(P) and they are I(1), how should I proceed? 1. generate a new variable on eviews by "generate Ed=E-d(E)" and "gen Pd=P-d(P)" where d=difference and use this as my center of analysis [would it be double differencing?] OR 2. just open the level variables and use them by inputting lag 1 in the VECM model, as VECM automatically differences them? Please help! Thank you very much!

[@yangchen5676]

it is so gorgeous! i think watching this video could save me at least 1 hour's lecture given by my rubbish lecturer

[@djepangkouamomoise2553]

how to integrate exogenous variable when you run up cointegration.

[@adi-px1gg]

hi, LondonPhD, can you write down the equation you made after you run the ECT

[@anshikasingh5822]

really very helpful video, thank you... :)

[@leoneltclemente]

in Eviews 7 is more simple. You need just do group of series, select cointegration, single equation, engle granger, and done. Run with with p-values.

[@thakersDotCom]

+Leonel Clemente ya and another simple way for rejecting/accepting Null is, check the p value and it is automatically given

[@fatihparlak7611]

How can I decide the number of cointegration equations in VECM model?

[@mznzulu]

what if the variables are I(2) can you use the Engle granger approach

[@libra142001]

Is necesary to have stationary variables to run a VAR model? help me!

[@thakersDotCom]

+Jonathan Chauca no

[@libra142001]

Why?? Muhammad Qadeer Thaker

[@thakersDotCom]

why don't you compare the adf values with critical values in the eviews table given. why are you comparing them with daviddson and mackninon values ? can you please explain.

[@alexanderwalmsley4447]

the crit values used on eviews are incorrect. as the t-values aren't distributed normally.

[@leoneltclemente]

+Muhammad Qadeer Thaker , you need t-tau published in mackinon for engle granger test, or use eviews 7 for automatic test.

[@thakersDotCom]

+Alexander Walmsley thanks dear.

[@thakersDotCom]

+Leonel Clemente thanks dear

[@vnought]

Very clear and helpful, thank you!

[@drewhc23]

Great explanation, thank you

[@ziddiilarka]

are you sure you are doing an engle granger cointegration test?

[@Mr104642]

you have to say that the procedure when the residual is not nonstationary.

[@yangchen5676]

he mentioned this before doing the first step.

[@efilimone]

thank you for the video

[@dalina-mariaandrei8752]

Thank you so much

[@jamie9898]

You legend

[@abdixaliim]

:) Merci