(EViews 10) Auto regressive Distributed Lag (ARDL) and ECM Model Estimation

[@Sarah17_]

since my data mostly not significant and need to re-run many times, i had a mental breakdown, worried how to run my data for the past few days before i found your video. thank you sir. may God bless you!

[@Laitnhoney]

Thank a Million Sir. This video has helped clear all confusions am having with ARDL model.

[@MAHADIHASAN-tp6sp]

thank you sir, this is by far the most informative video regarding ardl model ive found on youtube.

[@aec7787]

Thank you Best of Luck

[@HasanAlBanna-n4l]

Thank you so much sir. Very clear explanation.

[@HuyBui-zq2xh]

Thank you, sir for a very careful explanation. I have a question: at 21:43, for the level equation, why do we use "Case 3: Unrestricted Intercept, no trend"? Is it default to use Case 3?

[@entrepreneuriatrecherchesetcon]

Nice presentation

[@HuzMusFinanceData]

Very informative and helpful video... Thank you Sir... ❤

[@aec7787]

Most welcome

[@drashtichoudhart1129]

thank you, this vedio really healped a lot

[@dhakaramkadel3671]

The short-run data is not described. In which table is it? And what is the short-run effect of the independent to dependent variable?

[@ashveenaashveena570]

in ARDL model, after checking unit root test, do I have to use the differenced data or level one?

[@wondimumekonnen1769]

great

[@saumyamishra3176]

Thank sir for this tutorial. I have one query while finding the lag length criteria do we have to right all the variables in the endogenous variable box or just the dependent variable of our model..?

[@berhanugebre732]

thank you, it is a useful video for me. if you can post a related video.

[@hendelebiary1163]

Thank you so much sir for the constructive and helpful video. I just have one question, why in the conditional (unrestricted) ECM for bound testing and in the restricted ECM for the short run the E-views removes the last lag for each variable compared to the lag structure specified by the ARDL model? Thanks in advance for your help.

[@fitfirst4468]

yer ma, thats why

[@MaZZaNARMY]

Sir, in my eviews at the coefficients diagnostics part, i don't have the option to do Long run form and bounds Test, how do i get this option in my eviews?

[@zoyashah7826]

Which version of eviews u r using??

[@MaZZaNARMY]

@@zoyashah7826 eviews 10 , also my estimation setting when doing the ARDL shows PMG/ARDL instead of just ARDL like in this video

[@zoyashah7826]

What is PMG??

[@zoyashah7826]

Do you have cointegration and long run form option instead of long run and bound form??

[@MaZZaNARMY]

@@zoyashah7826 no i dont have that option at all

[@AyeshaSaqib-t8c]

If the value of ecm is not negative but statiscally significant and smaller than 1 what does that mean?

[@majeedhussain914]

it means that there is no adjustment towards the long run.

[@zoyashah7826]

Sir please clarify I have presented my research paper in a virtual conference..can I publish the same paper in a journal??

[@aec7787]

Hi Zoya...If the conference publishes your paper you can't republish it in another journal. Good Luck

[@drakulax52]

If data has a negative value I can to do log, what to do? variables are in percentage

[@majeedhussain914]

you can't convert data to the log.

[@omorfaruq5140]

Dear Sir, I want to estimate envirnmental kuznet curve, but when try to estimate ARDL then it show singular matrix, I add gdp per capita and square of gdp, how I can i solve the ARDL? please

[@guiltypleasure8379]

hey, sorry for intruding. but have you found the solution for this problem? i face the similar problem and idk what should i do. thanks in advance.

[@zoyashah7826]

Sir please tell me that for estimating long run coefficients we see the levels equation..but for estimating short run coefficients and to put it in a table for research paper,which equation to consider??

[@aec7787]

which version of E Views you use?version 10 you can find it directly from the outcome ECT the adjustment towards the long run .This is depend on the EViews version you use.

[@zoyashah7826]

Sir I am using Eviews version 10..but the problem is I have the option long run and bound test and not cointegration and long run option..

[@zoyashah7826]

And sir I have the option of estimating ECT but the problem here is of interpreting the results of short run as the results of all the variables with lag 3 is shown there.So how do we interpret that??

[@NoorulHijab]

@@zoyashah7826 dear I am using Eviews 9 .. after bound test when you go to coefficient diagnostic test.. different options will appear in which one will be cointegration and long run form

[@zoyashah7826]

@@NoorulHijab but I am not getting the option called cointegration and long run form..Instead of that I have the option as bounds test and long run form

[@harryshenry6258]

Thank you so much for the explainations and interpretations i would give a five star this video But my question is on the ECM regression table on the coefficient column What if you get the error correction term of -7.782080 and multiplied by 100?

[@aec7787]

Hi Harrys, sorry to reply late . First, check your data and run ECM again. The coefficient of ECT is negative and statistically significant which is good but you have a high negative coefficient please check your data and the software steps.

[@ramsharma8843]

Great video great mentor Please could anyone share references for steps of estimating ARDL model🙏

[@aec7787]

Hello Ram,See related videos on my channel step by step

[@johnabdulaijinapor1586]

How can one download the data for this video

[@usmansaleem1253]

Sir I have serial correlation in my ARDL results please guide

[@zoyashah7826]

You can drop that variable

[@zoyashah7826]

@@aec7787 thank u sir..I am from India.one thing I would like to mention that sometimes increasing the size of the sample doesn't remove the correlation problem as in my case..so I had to drop one of the variable .

[@aec7787]

@@zoyashah7826 Read another comment I mentioned how to remove the Autocorrelation problem from the model. Take the first difference for all variables and rerun the model again the serial correlation will remove. Do not forget to share and subscribes to my channel for more videos. Good luck

[@zoyashah7826]

@@aec7787 sir please guide me how to present short run coefficients on a single table under the ARDL model as the lag value is 3.

[@zoyashah7826]

Sir in eviews 10 how to find cointegration??

[@aec7787]

Zoa...sorry I did not get you, what do you mean that you can not find cointegration in EViews 10.If I understand your question you can run the variables at levels and run cointegration. or if you get the outcome of ECM and you did not find cointegration,, subscribe to mt channel and see more videos on cointegration.

[@zoyashah7826]

@@aec7787 sir I mean to say that in eviews version 10 there is no such option as"cointegration" to find the short run dynamics.To find the long run,there is an option of bound and long run in eviews 10 but not cointegration

[@usmansaleem1253]

Sir please do same in Urdu version for Asian students

[@aec7787]

Thank you Usman for watching the video. Sorry, I do not speak Urdu.

[@hafizm.sohail2012]

Watch Muhammad Saeed Meo videos (Urdu version)

[@usmansaleem1253]

@@hafizm.sohail2012 link please

[@aec7787]

Hi Usman, If you still have the problem of serial or Autocorrelation convert all the variables into the first difference then you will solve the problem. Do not forget to pass this video to your colleagues and friends and postgraduate students and even your professors and researchers for benefit of my channel and do not forget to share, subscribes, and comments for more videos. Best of luck

[@usmansaleem1253]

@@aec7787 thanks sir

[@wulan4607]

Hey sir I'm having a hard time when I try apply ARDL models .I did stability test but for QUSUM its shows the models is stable (the blue line didn't cross the red line ) but the QUSUM SQUARE, the blue line cross the red line lil bit . What do u think I should do or u can suggest ? 🥲