How to Estimate / apply and Interpret ARDL using Eviews

[@EzebunwoNyeche]

Thank you the explanation. I want to be clear wether we can use ARDL when all the variables are either stationary at level or are stationary at first difference.

[@econacademy16]

No

[@DineoJohane]

Thank you . Video lecture was very helpful.

[@aliahmadahmady2409]

Everything is clearly explained! Thank you very much sir for your great job.

[@Mmagnoliaa_]

Thank you for the great video!

[@angelaadomokhai4388]

Thanks for this explanation. Please i would like to know what it means to use unrestricted constant and unrestricted trend (case 5) and how to interpret having the constant and trend in the short run ECM but not having it in the long run. Also, after i am done running the long run and ECM tests, which of them do i run my post diagnostic tests on. Thanks in anticipation.

[@aam7486]

very clear, you deserve more subscribers

[@econacademy16]

Thanks

[@tomsingh2399]

Thanks for the explanation. Please it would help if you could explain the @expand section. When I put dummy variable in fixed regressors without @expand it works but with @expand it doesn't work. Is @expand for dummies mandatory?

[@parismoinas9569]

Thanks you it's really helpful. Just I would like to know what data we will use at level or first difference?

[@khagendraadhikari4796]

I need video about difference GMM and Sys GMM since you explain in very simple way.

[@econacademy16]

Noted

[@johnny7110]

Hello, I like your video, did you use raw data or rates? Another question I have is, when series being non-stationary at level but stationary at 1 difference, do I have to transform all data into 1 difference and run model after that or the stationarity test is just informative so I could run model with the data (raw or rates).. thanks much

[@econacademy16]

Stationary test is just for information

[@johnny7110]

@@econacademy16 thank you, can I use ARDL also when having different units of measurement in variables,? f.e. echange rates, inflation and GDP(in dollars),

[@nassims8022]

@@johnny7110 Hello thanks for the video and i have the same question please let me know if you found answer. can i use different unit?

[@birendranarayanshah]

Thank you for your informative tutorial. I think while running the ARDL model (for short-run relation) we must run the equation with the first difference rather than level data. Please correct me if I am wrong.

[@lucifergaming2229]

I saw that in many research papers but really I am very confused due to many researchers. Some are using level data while some are using differenced data. Can you please ellaborate. It would be so kind of you. Thank you.

[@philippetrape9295]

A great video! Thank you. I have just one question: The bounds test conducted shows that variables are co-integrated i.e. they converge towards long run equilibrium. Is there any error correction term (ECT) in ARDL models like in VEC models? And if yes where can we find it in Eviews?

[@rodrigoalvaradocantu562]

Great video and very helpful!! I have one question: Why im given coefficients for my dependent variable (GDPN in the examples shown) and how do i interpret them?

[@dhakaramkadel3671]

would you please see this? While conducting the ARDL model with four variables in Eviews, all the variables are non-stationary at level, but at the first difference, i.e., I(1), two variables, including the dependent variable, are stationary without taking their log, but two other independent variables are with log only. Should I need to use a log for all the variables in this situation?

[@emirarefa]

Hi, thank you for the explanation, I have a question, is it the same if i am using ols approach in estimating ardl model?

[@econacademy16]

Yes

[@emirarefa]

@@econacademy16 thankyou for the answer sir, but may i have another question, how to exactly choose the lag of dependent and regressor variable? I am recently trying different lags, as long as it is free from serial correlation and heteroskedasticity, and found out there is 2 option of lags that match with my preference, then how to choose between them? Is it the maximum or minimum one?

[@Midara_003]

Thank you so much 😊

[@snehalmishra1357]

Very well explained.

[@leehuilin9751]

Sir, when applying the ARDL model, do we need to run the data with log(logarithm) or without log?

[@charlesgodlovelamah9326]

it depends on the kind of data you have. if you are working with current value data, you will be in need to transform it into a logarithm, or if not you won't be in need. thanks

[@mosh71]

@@charlesgodlovelamah9326 what if the data has values less than 1 such as 0.5. It becomes as a negative number when doing logarithm. I have been wanting to know about this. thank you.

[@charlesgodlovelamah9326]

@@mosh71 In my opinion if the used value data contains negative sign you won’t be able to transform it because there is not logarithm of negative value, so in that case you will use the semi-logarithm. Having a negative sign after the transformation doesn’t matter, so you can keep doing your analysis. Thanks

[@dr.sureshmago9211]

Thanks sir for this valuable lecture. Please tell while applying ARDL model in panel data, other features of optimal regression equation like high r square and adjusted r square, Durbin-Watson value (near two), significance of Prob(F-statistic ), no serial correlation, No hetroscedasticity and normality of residuals is to be ensured or not need to be checked?? Regards

[@econacademy16]

No need to be checked

[@aam7486]

Thank you so much Mr, thank you

[@bbouchra1000]

Please, how to interpret the coefficients of the lagged explanatory variables. For exp GS(-1), GS(-2), GS(-") etc ?

[@zoyashah7826]

I also want to know the same thing..plzz help me

[@fidelmontenegro6195]

Hello bro, I am from Perú. Do the steps you are following also apply to an ARD model? You think I can contact you, I am doing a research work and I have doubts.

[@econacademy16]

[e-mail]

[@econacademy16]

You can contact me on [e-mail]

[@fidelmontenegro6195]

@@econacademy16 I sent you a message

[@denn2032]

Thank you. How do I incorporate the dummy variable in the ARDL? When I do as advised in the video I get a near singular matrix error.I will appreciate help

[@muilihamid8858]

You will include the dummy in the list fixed effect regressors

[@LifelongStudentBelgium]

if the f test in inconclusive what then?

[@kanikachawla3153]

sir after we got to know there is cointegration we need to run ecm. can u pls explain that also. how can i run ecm with dummy variable

[@fernnie7912]

Thank you so much sir!!

[@moizkhalid2910]

when i am going to apply ardl it shows asn error called (singular matrixs) ? fix it please.

[@econacademy16]

Increase number of observations or reduce number of lags

[@chimwemwekamaliza4145]

how to interpret the coefficients of the lagged explanatory variables in short ru

[@EsuyawkalNigusu]

how can i get ardl model on staata

[@LifelongStudentBelgium]

GDPN is gdp growth? so D(Log(GDP))?

[@dr.manjunathv6883]

Thank you so much sir

[@madinabunje3960]

Have a worry if at levels all variables are non stationary and in 1st differencing 2 variables are non stationary and also 2 variables are stationary can I use ARDL model? In addition I tested 2nd differencing it all came out stationary how then can I process my analysis?🙁

[@econacademy16]

In this case you can not use ardl.. Check cointegration and read papers on var, vecm and ecm. It will help you

[@donasp5391]

IF T=10 (years) and N= 350 companies, could I use ARDL?

[@snehalmishra1357]

How to estimate cointegrating equation?

[@fitfirst4468]

Teach me how to ADRL !

[@moizkhalid2910]

hello, how could i find Variable's Data about more than period of 20 year e.g (ROE,CAR,FDI )? kindly guide me the link to find out data according to my region is (paksitan) thank you

[@snehalmishra1357]

How to estimate granger causality?

[@محمديوسف-ل6ع2خ]

THANK S

[@lucifergaming2229]

Hello Dr. Wali, nice informative video. I have some questions regarding ARDL eviews. If you can provide me your email address. It will be much easier for me to ask through email. Thank you.

[@econacademy16]

[e-mail]