(Stata13): Panel ARDL Estimations (Steps 9 & 10)

[@CrunchEconometrix]

KZbin recently changed the way my content will be monetised. My channel now needs 1,000 subscribers. So it would be amazing if you show your support by both watching my videos and subscribing to my channel if you haven’t done so already. Monetising my videos allows me to invest back into the channel with some new equipment so this small gesture from you will be extremely huge for me. Many thanks for your support….CrunchEconometrix loves to teach, support my Channel with your subscription and sharing my videos with your cohorts.

[@dr.elkhanrichardzada9855]

I meant bi-directional causation (Panel Granger causality test for PMG ARDL)

[@CrunchEconometrix]

@@dr.elkhanrichardzada9855 Not sure if testing for bi-directional causality will make any intuitive sense in panel ARDL since that will imply using each of the regressors as a depvar.

[@sarahahmedchawsheen5455]

Thanks to you Dr. Adeleya for the comprehensive videos, they are all very helpful to me.

[@CrunchEconometrix]

Thanks, Sarah for the encouraging feedback. Deeply appreciated 🙏🥰

[@mauriceligulu6782]

Thanks so much. Your tutorials have helped me improve so much in econometrics. you have made it simple.

[@CrunchEconometrix]

Thanks, Maurice for the encouraging feedback. Deeply appreciated!

[@mishalkhaled8327]

the best lectures that I watched ..thanks so much Dr Ngozi

[@CrunchEconometrix]

U're welcome, Jasem! 💕

[@mishalkhaled8327]

honestly, I am learning a lot from you@@CrunchEconometrix

[@CrunchEconometrix]

@@mishalkhaled8327 Good to hear this, Jasem. Glad to be of help. Kindly tell others too by sharing my Channel link with your students and academic networks...thanks!

[@solomonbola2493]

May God continue to bless you...very educative and insightful...thank you ma

[@CrunchEconometrix]

Same to you, Sir...and thanks for the encouraging feedback. Deeply appreciated. Kindly share the link to my Channel with your academic network in Nigeria and beyond. God bless.

[@solomonbola2493]

@@CrunchEconometrix Okay ma

[@mishalkhaled8327]

Thank you so much I recommend this for all people

[@CrunchEconometrix]

Thanks for the recommendation, Jasem! Please tell others too😀

[@sonu4ever]

Hi Madam, you did really a good job

[@CrunchEconometrix]

Thanks a lot, Sir!

[@yassinyahia2453]

Thumb up for u professor for ur great job....I would like to know if u have videos on gravity model?

[@CrunchEconometrix]

Oh not at all...😊

[@wilfriednguie6879]

hello professor, i would like you to make a video or show you how to do the Westerlund (2007) uniatary root test please

[@CrunchEconometrix]

Suggestion is noted, Wilfried. Thanks.

[@m.walidhemat6319]

how to set the commad for lag order of (1111)? could you show in your example. I am conducting panel ardl . The result of the Stata is quite different than that of the Eveiw. Maybe there is problem in right use of lags.

[@CrunchEconometrix]

I'm familiar with 1 0 0 0 lags, which is what I showed in the video. Not surprising that Stata and EViews produce different results. It happens sometimes.

[@taallahlilia]

Thanks a lot Dr. Nigozi, may God continue to bless you, it is very helpful. But I have a question: I had an issue with the diagnostic tests, if you can guide me for a video to see how I can perform them to my panel data ! and do I have to perform N numbers of diagnostic test for each country ?

[@CrunchEconometrix]

Hi Taallah, you may need to check papers that used panel ARDL on the diagnostics to engage. Thanks.

[@nanakwartengchannel6852]

Hello Dr., thanks for the tutorials. Please just a quick one; from which section can we find the do-file for this tutorial on your website?. Thanks

[@CrunchEconometrix]

Hi Nana. I deactivated the link to my Google drive since 2019 due to abuse and unethical conduct. Since then all Stata dofiles used in my videos are no longer free but available on my website upon payment. Here's the link cruncheconometrix.com/view/datashop.php

[@mahinurmimi9782]

Is the causality test is same for time serious variable Ardl model?

[@CrunchEconometrix]

Hi Mahinur, I'm unable to understand your question. Kindly rephrase.

[@moviesanddramakorea]

my pmg analysis for ech country is not showing the full data instead it gives me just ECT and constant

[@CrunchEconometrix]

Hi Kezia, I am not sure why that is. You may want to post your query to Statalist.org for more constructive feedback. Thanks

[@lehuy7279]

I have a data with T=11 < N=63, and i want to estimate the causality relationship in short run and long run. Is panel ARDL suitable?

[@CrunchEconometrix]

I mentioned that panel ARDL is only suitable for N < T panel structure. Kindly watch the introductory video b4 the rest of the videos in the series. Tx

[@KeziaSpeaksHerMind]

Professor, you talked about that if we want to run diagonistic tests for the Panel ARDL, we have to perform a time series analysis for each group(country). Does this mean that we can perform any time series analysis(linear regression e.t.c) or we have to perform time series ARDL analysis?

[@CrunchEconometrix]

Beyond what I said I showed exactly what to do. Kindly watch the clip again, thanks.

[@eyuptanil4285]

can you share Diagnostic (optimal) video links here, ı cant found your video list

[@CrunchEconometrix]

Hi Eyup, if I have such you will find it on my Channel. Kindly browse within. Thanks.

[@eyuptanil4285]

@@CrunchEconometrix ı would like to ask o questioans about of my master degree thesis. ı am researchin EU agriculture subsidies. ı have 4 subsideis variables to examine how to effect total inputs costs. I have N:25 T:15 balanced panel data. my some variabels I(0) , same of I(1). My CD test results , the series hava serial collerations. and my homojenity test results; series is heterejen. ı couldnt apply Panel ARDL because of T is small. ı cant expand T, because of some important limitians. Is GMM is best alternative for my work? which methods can ı use to anlysis this it.?

[@CrunchEconometrix]

Eyup, you can apply any technique suitable for short panels (N>T).

[@eyuptanil4285]

@@CrunchEconometrix profoser how can ı apply Fourier Panel KPSS test? ı cant found stata codes

[@moviesanddramakorea]

Ma, how can one perform an FMOLS test on Stata

[@CrunchEconometrix]

Kezia, type "help fmols" in the Command Window for Stata prompts and guide.

[@moussaconde9309]

Hello Teacher, I want to Ask about pmg result import on overlaef only Long run case . I have Seen your tuto on KZbin. Sincerely yours

[@CrunchEconometrix]

Hi Moussa, I could not understand your statement. Kindly rephrase. Thanks

[@moussaconde9309]

@@CrunchEconometrix I meant that I want to import the results of my Stata regression of the PMG model on latex while only keeping the part relating to long-term estimates.

[@CrunchEconometrix]

Hi Moussa, I have no idea about latex. Never used it. You may want to check out other online resources for more information. Thanks

[@ranjanachaudhary7907]

Hello prof. Thank you for the most helpful video . i have one question professor can we use pmg dfe fixed effect and random effect regression at the same time

[@CrunchEconometrix]

Ranjana, those techniques apply to panel data with different structure. I explained this in my panel videos. Kindly watch.

[@ranjanachaudhary7907]

@@CrunchEconometrix Thank you so much but can i get the link please it will help!

[@CrunchEconometrix]

Kindly search through my panel data videos.

[@kanchandatta4668]

Mam mg estimation is not coming. Pmg and dfe is coming but mg estimation is not. How this problem can be solved. All the commands are followed but still not coming. Hence we are not comparing pmg and mg or mg and dfe etc. please guide

[@CrunchEconometrix]

Hi Kanchan, perhaps it is due to Stata updates. Otherwise, you should have "mg" just as I showed in my videos.

[@KFP385]

Dr Ngozi, I tried to access the do files on your website but paystack doesn't seem to work. Is there any other way I could obtain the do file with the dummy variables?

[@CrunchEconometrix]

Hi TLus, I just responded to your email. The payment channel is functional. Sorry for the challenges. Kindly try again. Thanks

[@btessa2dc]

Dr. Ngozi, Do you have any references about test of causality to support your arguments in this video? I want to add this analysis to my research paper, therefore will need to support that with some references. I have reference when the causality test is perform using time series data, but in my case I am using panel data, and believe that the same approach may not hold. Thanks

[@CrunchEconometrix]

I indicated some references at the end of the videos. You can read them in addition to several papers that used the PMG/MG to back up your analysis.

[@mabrukamohamed83]

thanks a lot .please do a video about NARDL

[@CrunchEconometrix]

I will, definitely...thanks Mabruka! Kindly tell others about my Channel too...

[@mabrukamohamed83]

thank you for sharing your knowledge with us

[@CrunchEconometrix]

@@mabrukamohamed83 My humble pleasure, Sir!

[@bakytzhanzhaparov1846]

Hi, professor. Can we argue that causality runs from x to y? Or there maybe reverse causality as well? In your case, can we say that credit negatively affects economic growth and not economic growth has negative impact on credit? Is causality in panel ardl is unidiriectional if results are statistically significant?

[@bakytzhanzhaparov1846]

I am writing dissertation and my supervisor asks me to solve possible bi directional causality, what can I do? Or panel ardl is enough to argue uni directional causality? Sorry too many questions from me, but I am writing a dissertation and it is very important for me

[@CrunchEconometrix]

Bakytzhan, I provide references at the end of my videos and there are several articles/online resources about CAUSALITY. Read up. I have always maintained that video tutorials are insufficient. Solidify your understanding by reading the extant literature on causality. Thanks.

[@dorstellgh6238]

Hello Dr, please how can i determine R squared and Adjusted in ARDL Estimations. Thanks

[@CrunchEconometrix]

Hi Nana, please always clarify which ARDL is referred to. Panel or time series?

[@dorstellgh6238]

Panel ARDL

[@CrunchEconometrix]

@@dorstellgh6238 R2 (not adjusted) is available for DFE (not PMG) when you export the output.

[@dorstellgh6238]

Noted well Dr. thanks

[@AhmedAhmed-jv4xe]

it is the same steps for panel nardl?

[@CrunchEconometrix]

Yes Ahmed...the video series is on panel ARDL. May I know from where (location) you are reaching me?

[@AhmedAhmed-jv4xe]

@@CrunchEconometrix thanks for your answer Mrs I asked about the steps for A nonlinear Panel ARDL approach

[@CrunchEconometrix]

@@AhmedAhmed-jv4xe No idea, at the moment.

[@gayatrigogoi571]

how can we know that our model need to use panel ARDL because we can also run the simply panel regression model.please do reply

[@CrunchEconometrix]

Hi Gayatri, the introductory video "Basics of Panel ARDL" has the background information to your query. Kindly search through my Playlist and watch. Thanks! May I know from where (location) you are reaching me?

[@haddaderadra2310]

thank you, please i have a question : is it possible to do diagnostic test to the panel or it is only for individual units ?

[@CrunchEconometrix]

But I mentioned that diagnostics are optional.

[@haddaderadra2310]

​@@CrunchEconometrix yes, thank you

[@nawalin12]

@@CrunchEconometrix Helo Dr... See you again here.. If that the case without any diagnostic test, is the result still valid (academically accepted? what is the justification? If we perform, so that means we have to perform N numbers of diagnostic test for each country?, if there is any country that not fulfilled the diagnostic test, then how ? Thank You ..

[@nawalin12]

I used STATA 16, how to write lag specification in the command, because the command do not have any room for lag specification as in time series ARDL, but when I compare with eviews, the output in Stata is same like output Eviews with lag 1 for both dependent and independent variable. I have tried to run lag determination following your command but, what I wonder is how to plug / insert in xtpmg command, if i put lag then error. Thank You

[@taallahlilia]

@@nawalin12 i have the same question

[@ahmadmughal4079]

mam thanks for sharing worthy knowledge . my question how to check short run or long run two way causality? and second how to perform diagnostic test? please send me commands thanks

[@CrunchEconometrix]

Hi Ahmad, apologies for the late response. Please watch all my CAUSALITY videos for better understanding and note that dofiles are available on my website at a token fee. Kindly visit the STORE www.cruncheconometrix.com.ng/shop/ to make the purchase, thanks.

[@nadahusseinshokry5886]

Hi Adeleye, it’s me again asking 😄 So, I used the loop command you wrote for ardl lag selection on unbalanced data of more than 35,000 panels and 142 time units. The problem is an error keeps appearing: “no observations”. It is a very persistent error, do you have any idea why would such an error occur?

[@CrunchEconometrix]

Hi Nada, if you had watch the 1st video on the basics of panel ARDL you would have known that your data must be in N

[@nadahusseinshokry5886]

CrunchEconometrix great thanks. I did watch it but I didn’t notice. Do you know what to use if my N>T?

[@nadahusseinshokry5886]

Do you think GMM is suitable with such a long panel?

[@CrunchEconometrix]

@@nadahusseinshokry5886 No dear, it's designed for short panel. May I know from where you are reaching me?

[@CrunchEconometrix]

@@nadahusseinshokry5886 Please watch the 1st video as advised. GMM is designed for short panels.

[@nssofod]

Madam hi. It's Debasis. Thanks for your videos. I hv two queries 1. how to analyze the results of cross section dependence test in panel data? 2. What to be done if it rejectsc the null hypothesis of independence? Plz share if you hv any video on that.

[@CrunchEconometrix]

Hi Debasis, I have noted these queries. Added to my videos-to-create list. Thanks!

[@nssofod]

@@CrunchEconometrix ok mam. Thanks. Hv a nice day

[@nssofod]

Good morning mam. What should I do when my westerlund panel cointegartion test gives "no cointegartion" result? Is there any other alternative? Or how do I address this result?

[@CrunchEconometrix]

Try the Pedroni test.

[@nssofod]

@@CrunchEconometrix you pedroni test for panel cointegartion? It has been done and significant result. But many papers argue that pedroni cannot give efficient result in presence of CSD. We hv to do westerlund.

[@ty3454]

Can I have the do file for this panel causality test?

[@CrunchEconometrix]

Hi Yu, I don't have a separate dofile for that but you can infer causality from the statistical significance of the coefficients. Watch any of my causality videos for guidance. They are quite explanatory.

[@dr.elkhanrichardzada9855]

@@CrunchEconometrix and what about reverse causality?

[@CrunchEconometrix]

@@dr.elkhanrichardzada9855 I stumbled on this response on ResearchGate, which I often apply in my work. You may find it useful, though I never used it for panel ARDL analysis: "In order to check whether your regression suffers from reverse causality, the best solution is to re-estimate the model using a lagged IV so that the DV(t) is a function of IV(t-1): the logic is that if there is no reverse causality the link between the IV and the DV has to be the same irrespective of the time lag. Thus, if the sign of the IV changes and is significant, this means that your regression suffers from reverse causality; if there are no changes in the sign of the IV, then you can rule out reverse causality". Hope this helps.

[@dennisbaidoo5995]

Great

[@CrunchEconometrix]

Thanks!

[@busaritajudeen9168]

You are amazing teacher. When are you going to produce Random Effect and Fixed Effect on Panel data with STATA? You be Amazing mother too

[@CrunchEconometrix]

Thanks for the kind words, Busari. I do have those videos. Check out the Stata Panel Data Playlists. Thanks 😊

[@busaritajudeen9168]

Thanks for quick response. What i am trying to say is "Panel Data Analysis with Stata: Fixed Effects and Random Effects Models" It wasn't treated in Stata playlists

[@CrunchEconometrix]

@@busaritajudeen9168 OOps! my apologies. I treated them under "Hausman Test". It's in the Playlist. Sorry about the mix-up.