Johansen Cointegration Test. Model One. STATA

Data to reproduce model:
docs.google.co...
KZbin: / sayedhossain23
Facebook: / 393927400736679

[@Paktvlounge]

I would first like to thank Sayed Hossain for his continuous effort and dedication to this subject and the profession of teaching. To all those people who have a problem with Sayed Hossain's English and rate of speech can simply go and look elsewhere. His pace might take more time but makes it much easier to understand the crux of it all.

[@sayedhossain23]

Dear Mohsin, 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/

[@yosriouni]

Tip: Put the speed of the video at 1.5

[@sayedhossain23]

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/

[@TheSergiolion]

Darth Vader is talking to me....

[@synncmaster]

best advice

[@eiz8745]

This vid indeed is very helpful. but I put the speed at 2 lol

[@yulu7307]

i run it at 2

[@musicadidi1014]

Hi Mr Hossain ! Really usueful videos for researchers and students. Thanks a lot ! from Colombia

[@sayedhossain23]

Liliana Gomez Thank you. I would like to invite you to join Hossain Academy Facebook for greater interaction about economics, finance and econometrics. Thank you Sayed Hossain from Hossain Academyfacebook.com/groups/hossainacademy/

[@imrityaaqub3566]

A very useful video! You provide very clear and good explanation. Well done and thanks a lot. From Paris

[@sayedhossain23]

+imrit yaaqub 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/

[@sayedhossain23]

1. There is no guideline in this regard so far I know. But the standard lag selection is 2 and that is why I have taken two but few suggests to take more than 2 lag to check. 2. Guidline is: when you fail to reject null hypothesis, you will have to accept the corresponding null hypothesis. You can not proceed further. Suppose, in Eviews, At most 2, is the null hypothesis and if you fail to reject it, meaning that there will be 2 cointegration equation.

[@TaiChenTaiChen]

Sir, Thanks for your video. Time series becomes more easily understood from your video.

[@sayedhossain23]

+Tai Chen 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/

[@TaiChenTaiChen]

+Sayed Hossain Thank you for your invitation. It is my pleasure to joint your facebook group.

[@yeshiwasewinetu3295]

I really like very much you vedio ,keep it up pls!!

[@sayedhossain23]

Suppose lag selection criterion or method has advised me to take lag 2 and that is why I have taken lag 2. Normally Constant is best option and used widely as that fits most of the data.

[@Danwell86]

Hi Mengru, I am unsure what data you are using, but one common problem is the use of trading days'. This obviously skips over weekends and holidays, therefore the time series has gaps. One solutions (which is a bit of a cheat) is to change your time variable to 'periods'. Stata will see this as a temporally-consistent time series with no gaps. I.e. in my sample is use days 1 to 6400. Hope this helps.

[@synncmaster]

Thanks, your videos are saving my dissertation

[@sayedhossain23]

Thank you. I would like to invite you to join Hossain Academy Facebook Group (Data Analysis) at below link and join our group discussion about modelling. Thank you once again, Sayed Hossain from Hossain Academy. facebook.com/groups/hossainacademy/

[@dr.m.serajurrasul2833]

You talk slowly but it is helpful for beginners to understand the videos.

[@thetruth4712]

Thank you for the video. BTW, what does LL and parms refer to and how can we interprete the value of the LL. Regards.

[@BagasSept]

I'll be nuts the time this test is done. Nonetheless you did a good job. But what should I do now that my data have no co-integration..??

[@sayedhossain23]

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

[@MarizaMontesdeO]

Dear Sayed I found your videos very useful. Thank you very much. I have some questions. Why do you take your lags to be 2? and why do you chose "constant"? Thank you very much!!!

[@kanyakamysti]

What if after converting them into first differenced they are still non stationary, then?

[@shuaiyuan2164]

Dear Mr Sayed, I would like to say thank you to you because your video has helped me a lot with my dissertation. And could I ask a stupid question: if a variable is stationary at its level and is also stationary at its first difference, could this variable still be included in the Johansen Cointegration Test and be examined the cointegration with other variables which may be I(1)? Thanks!

[@VarunChotia]

Hi Sayed. Thanks a lot for this video..Proves to be very useful !! Couple of questions on this : 1) Why lag of 2 has been taken.. this questions has been asked previously also but i did not completely understood your reply to this ? 2) When we check for 1 COINTEGRATION and we get to know from both the tests that there is 1 cointegration, then why we stop then only. Why we do not go forward and check for 2,3 cointegrations because if we check , we will see that there is 2 and 3 cointegrations

[@yaw258]

Hi I appreciate your videos on cointegration using stata, but i cannot download the videos as i did with the eviews presentation. Why Thanks

[@gayatrigogoi571]

How to do panel fmols in stata?

[@forooghfarzami6010]

I always use your videos and they are really useful , thank you for sharing your knowledge Sir.

[@sayedhossain23]

You are welcome

[@sayedhossain23]

Thanks Madam. Why not you join our Hossain Academy Facebook club where we are chatting about econometrics? If you want to follow, follow following link below. Thanks. Sayed Hossain facebook.com/groups/hossainacademy/

[@amaranwar1249]

Hello Hussian, I get this message, what is the solution: cannot fit model because of collinearity

[@sayedhossain23]

***** Model fitting deals with R square, If there is collinearity problem in the model, it makes the significant variable insignificant.

[@amaranwar1249]

Sayed Hossain What's the solution? I mean how to deal with collinearity before checking Johansen Cointegration Test.

[@sayedhossain23]

I have sent you an invitation. Would u mind to accept it and join our newly built community. Thank you Sayed Hossain from Hossain Academy

[@amaranwar1249]

Sayed Hossain I did that Hussain. Could you kindly reply to my question: What's the solution? I mean how to deal with collinearity before checking Johansen Cointegration Test. 

[@sayedhossain23]

In Johansen test, you do not need to check collinearity. The precondition of Johansen is all variables should be I(1). Only then you can run Johansen.

[@felixagyei-sasu1039]

sir what if the model shows that we should fail to accept the null hypothesis of zero co-integration. Also if the hypothesis for one co-integration is failed to be rejected as well.

[@sayedhossain23]

Fail to rejecting null hypothesis of zero means there is no cointegration among variables so you can not run VECM model but can run VAR model only. If the hypothesis for rejecting one cointegartion is failed meaning that there is one cointegration meaning all variables are cointegtaed, so you can run VECM model.

[@felixagyei-sasu1039]

Thank you sir.

[@mohammedmahinuralam2796]

Dear Mr./Dr. Sayed Hossain! Thank you very much indeed for sharing these useful videos for interested students and researchers in applied econometrics. I feel truly grateful to you for your service to KZbin learning community to which I happily belong. I am wondering if you please could share real examples from your own research papers in addition to the great job you are already doing for us, if time and energy permits :)

[@sayedhossain23]

Mohammed Mahinur Alam Dear Alam, Thank you. Please join Hossain Academy Facebook below for greater interaction with me regarding data analysis. Thank you Sayed Hossain from Hossain Academy facebook.com/groups/hossainacademy/

[@sinchet]

Hi Sir. Thanks for your video, it has greatly helped alot for my regression. However, I have a question here, do we use the level data or the first-differenced data to perform the co-integration?

[@sayedhossain23]

If you do Johanen Test of cointegration, data must be at level meaning that nonstationary...

[@Ibi252]

Hello! Nice Video but what was your criteria to choose lags? Why you opted for two lags why not 3 or 4 lags? I have two daily time series what lag should I choose?

[@sayedhossain23]

+Zubair Ali Raja 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/

[@Ibi252]

I have already joined you there :)

[@sayedhossain23]

+Zubair Ali Raja Please post your question there

[@nguyentuyen5139]

Dear Sir, What happens if the time series are not stationary at the first difference but the second? How to check cointegration between X and Y when they are stationary at the second difference? Thank you!

[@sayedhossain23]

Dear Nguyen, Thank you. I would like to invite 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/

[@sayedhossain23]

Sorry do not know about it what is happening there

[@hasyaamana2985]

Hi, thank you for the video it really helps! Im doing a time series analysis as well, but all of my variables are stationary at level. how can I do JJ test for cointegration in this case?

[@sayedhossain23]

Thank you. I would like to invite you to join Hossain Academy Facebook Group (Data Analysis) at below link and join our group discussion about modelling. Thank you once again, Sayed Hossain from Hossain Academy. facebook.com/groups/hossainacademy/

[@rupaltak3475]

Sir...what will we do...when ..trace statistic value at 0 rank is greater in 1st table N in second max stats is less than critical value at 0 rank

[@rupaltak3475]

Please answer

[@nadirshah3870]

Can you please give me the Data which u have used here so that i can do practice the test using the same data set?

[@sayedhossain23]

Dear Shah, Thank you. I would like to invite 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/

[@mengruzhou7815]

Hi, i am very thanks you to do this stata video. it is very useful, but when i do the test it says the sample has gaps r(498); and  invalid name r(198); could you help me？thanks!

[@Mbodou22]

Hello ! I want to know how to do a cointegration in panel data ? Thanks

[@sayedhossain23]

DearAdji, 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/

[@federicolosa1302]

Thank you, very helpfull. But how do I get the value of the cointegration vector in STATA?

[@sayedhossain23]

Dear Losa, Thank you. I would like to invite 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/

[@thakersDotCom]

sir your videos are very useful to me. i am working with time series data in my dessertation. i want to ask that shall we check cointegration test before running VAR i mean if our data passed cointegration test then shall we not go for VAR ?

[@sayedhossain23]

First Johansen Cointegration test to know whether VAR or VECM model to be used....

[@thakersDotCom]

if it passed johanson cointegration test then shall we go for VAR or not ?

[@sayedhossain23]

Please see my few videos to check when u should run VAR or VECM..

[@sayedhossain23]

Yes. If Johansen approves it, go for VECM

[@tinhnguyen-ok9dd]

Hello sir, in case datas become stationary mixture of I(1) and I(0), i can not use Johansen to test Cointegration. I must use ARDL to test cointegration. Can you help me how to test ARDL on stata.

[@sayedhossain23]

tình nguyễn I have never tried with STATA but can be possible. I shall do it may be in future

[@ryanwilson9283]

what am i not doing right to get different ranks With max and trace statistics the former usually beining one rank less and sometimes even if trace statistics shows a 1 cointegration max stat fails to reject at 0 rank

[@sayedhossain23]

Ryan Wilson Yes both can give different result. In that case you need to choose any one as benchmark to take decision. But it would be good when both gives the same decision.

[@arturox1997]

what if i get no star

[@sayedhossain23]

Thank you. I would like to invite you to join Hossain Academy Facebook Group at below link and join our group discussion. Thank you. Sayed Hossain from Hossain Academy. facebook.com/groups/hossainacademy/

[@ibrahimahmad5161]

Thank you once again for the effort. Please what will i conclude if in all the the ranks my trace statistics is more than 5% critical value? unlike your result in which at rank 1 trace value is statistics less than 5% critical value. please i need your help Sir. from UK

[@sabuhiisazada815]

big thanks,sir

[@limryuzaki7716]

Thank you so much

[@thejuanchc100]

THANKS VERY USEFULL

[@hanwooljang6460]

Thanks alot~!

[@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/

[@diegovillacreses7071]

Dude you talk so slowly that I'm gonna die!

[@sayedhossain23]

+Diego Villacreses Sorry for it. good luck

[@diegovillacreses7071]

+Sayed Hossain sorry for the hard words. This is actually really useful. Thanks!

[@azatzhonkabayev638]

Agree+++

[@shakibishfaq8627]

This is why you speed up the video.

[@ibossox]

Even 2x speed is still time-consuming......

[@dhruve4c]

You need to speak faster

[@sayedhossain23]

You are right, Next time I will,

[@dhruve4c]

Or learn English init

[@archstevej]

This is so dang rude, his speech is nice and clear, making it easy to follow.