(Stata16): Heteroskedasticity and Robust Standard Errors

[@CrunchEconometrix]

I want to appreciate all my subscribers from across the globe (Africa, Asia, Europe, the Middle East, The Americas, and The Pacific). Thank you all for your support. I am encouraged by your comments, questions, likes and critiques. They keep me focussed and poised to do better. I will continue to contribute my little quota such that every student and researcher will independently analyse his/her data. My teaching approach is very practical. I adopt a do-as-I-do style. Many thanks to those who have supported me by telling others. Once again, CrunchEconometrix loves to teach, support my Channel with your subscription, likes, feedbacks and sharing my videos with your cohorts. Please do not keep me to yourself (lol) inform your friends, students and academic networks about my Channel. Tell them CrunchEconometrix breaks down the econometric jargons and teaches with simplicity. Follow me on Facebook, Twitter and Reddit. Love you all, greatly!!! 

[@CrunchEconometrix]

Hi Jovan, thanks for the positive feedback. Deeply appreciated! Unfortunately, I am unfamiliar with those software. Regards.

[@uchennamagnus8522]

YOU'RE A REAL BLESSING TO YOUR GENERATION. PLEASE, KEEP DOING THIS GREAT WORK.

[@CrunchEconometrix]

Thanks Uchenna for the encouragement. Gracias!

[@dulithapeiris9727]

madam you are one heck of a lecture, thank you very much for genuinely sharing your knowledge ive been learning many things and applying the knowledge in my own researches ...you are the best..keep up the good work

[@CrunchEconometrix]

I'm humbled by your encouraging feedback, Dulitha. Deeply appreciated! Please may I know from where (location) you are reaching me?

[@GustavoRodrigues-of1hl]

You already saved my life once in a last exam. This time I come to real learn and you make that so easy and clear! Mixing the theory with the applications in softwares. You are a real queen!

[@CrunchEconometrix]

Hahahaha, great to hear...Gustavo! Thanks for the compliments. Humbly taken.

[@josephlanre]

Thanks crunch econometrics. You videos have been a revelation. It has helped me in my online econometrics train

[@CrunchEconometrix]

Thanks Joseph, for the encouraging feedback. Deeply appreciated! Please may I know from where (location) you are reaching me?

[@josephlanre]

@@CrunchEconometrix Ibadan. Nigerian Institute of Social and Economic Research

[@bendiciondelpino2965]

Dear Adeleye, your channel has been really helpful for my research, thanks !!! Best wishes 4 u!! 👏👏👍👍

[@CrunchEconometrix]

You're so welcome, Ben!

[@adedoyinlawal5396]

Its great watching your lecture.

[@CrunchEconometrix]

Hi Dr. Lawal, thanks for the encouraging feedback. Deeply appreciated😊! I will appreciate it if you can help share the link to my KZbin Channel with your students and academic community🙏

[@nabeelabhatti5749]

I want to appreciate your effort 👍. Can you please upload the video of logit model especially in case where all the variables are categorical and some variables for example women education have 5 categories.

[@CrunchEconometrix]

Hi Nabeela, thanks for the suggestion on the logistic regression. Noted. I will make the videos once I fully understand the technique. Kind regards😊

[@hamanmahamataddi8600]

Thank you Mam for your endeavours to help us be familiar with econometrics. I would like to know if you have some video on how to perform panel bootstrap causality (konya,2006). Best regards.

[@CrunchEconometrix]

Oh not at all. But I have noted it down.

[@argandefar2535]

really, I appreciate you.from ethiopia

[@CrunchEconometrix]

@@argandefar2535 U're very welcome. Kindly share the link to my KZbin Channel with your students and colleagues in Ethiopia...thanks!

[@aminulkarimakash7942]

Dear Professor, your lectures are outstanding, I found all the videos very easy to understand but sometimes I find it difficult to categorize the videos because it is not serially mentioned. Can you please order all the videos accordingly? TIA

[@CrunchEconometrix]

Thanks Aminul for the positive feedback. Deeply appreciated. You can either follow the guides on the prerequisite videos to watch or SEARCH through the 9 playlists. My videos are well sorted. Thanks.

[@keshassom]

Hello. Thanks for video. Why don't we ever have a video for example times series analyses or a playlist for it. For the first step to the last one. Another playlist for panel etc. The most critical problem of data analysis is how to manage datas. The model specification is not well showed. We don't even know how to remove a trend in a serie. Even in the dickey fuller test, there is no video which apply exactly the box and Jenkings methodology. You may think about it. Kinds regards

[@CrunchEconometrix]

Thanks Kesha, all points noted. My videos are sorted into 9 Playlists. You may need to browse through to find what you need. Regards.

[@yousafkhankhalil8726]

Thank you very much for your great tutorials

[@CrunchEconometrix]

Thanks for the encouraging feedback, Yousaf!

[@chandlerinman4393]

Great video! To the point and very clear

[@CrunchEconometrix]

Thanks Chandler, for the encouraging words and feedback. Deeply appreciated!

[@mohammadhanif6589]

This is very helpful. Thank you very much

[@CrunchEconometrix]

Glad it was helpful, Mohd!

[@gilbertleon3080]

Dear Professor, your lectures have helped me a lot in understanding econometrics. It has been a very tricky subject. I have one request, could you possibly explain a non-linear causality test please?

[@CrunchEconometrix]

Hi Gilbert, I am humbled and encouraged by your comments and positive feedback. Deeply appreciated! As to your query, non-linearity will firstly be defined from your model specification. Eg including the square of X as a regressor. When Xsq and Y are subjected to causal analysis you have either the existence or lack of non-linear causality. Please may I know from where (location) you are reaching me?

[@sayantikapaul2920]

Thank you so much ma'am you always make stata very easy to understand

[@CrunchEconometrix]

Thanks Sayantika for the encouraging feedback. Deeply appreciated!

[@rochnaarora7478]

You are doing a great work.. Its a humble request if you can provide a tutorial on "hurlin venet causality" (2001,2004,2005) (if you have heared or read anything about it) process that is an extension of granger causality mechanism.. It would really be of great help. Looking forwrd to ur reply

[@CrunchEconometrix]

Hi Rochna, I'm humbled by your positive feedback...and grateful for the Causality papers. I'll look them up and if I understand the procedure will definitely create videos from there. Please may I know from where (location) you are reaching me? Thanks.

[@rochnaarora7478]

@@CrunchEconometrix India

[@CrunchEconometrix]

@@rochnaarora7478 Awesome! Kindly share my videos and KZbin Channel link with your colleagues. They need to know about my Channel and may God bless you as you do...thanks!

[@z.b00t]

Dr please why don’t you explain the results so we get a better understanding? Please do that for us

[@CrunchEconometrix]

Results interpretation is generic. I have videos on how to interpret results. Kindly check out my playlist. Thanks.

[@z.b00t]

Thank you

[@toyinkezia5181]

Hello Ma, I came from Sis Olabisi wall. I have been following your videos and I must say, it has been helpful so far. Ma please can you be attaching the links to sequential classes so that it would be easy to find. Thanks

[@CrunchEconometrix]

Hi Toyin, thanks for the encouraging feedback. Deeply appreciated! You can always browse through my Channel to locate all the videos. Thanks.

[@mym2837]

hi Miss...Miss make a video on kendall tau and spearman rank correlation

[@CrunchEconometrix]

Thanks for the suggestion. Well noted.

[@tegegnnigussie1873]

Hi Dr can you give me some clues about DASP software package. Especially how ti install on stata.

[@CrunchEconometrix]

I have nit used the DASG software, So, no idea at the moment.

[@mamadouniambele4176]

Thank you so much!!!

[@CrunchEconometrix]

You are welcome, Mamadou!

[@김대은-e5d]

I have a question, I already used Robust Standard Errors but the heteroskedasticity test still indicated heteroskedasticity. Is there anything I can do to solve this? One of my academic advisor said that if I already utilized robust standard error, the test would be valid regardless of the heteroskedasticity test , is this true? P.S. Your videos are super helpful, thank you very much.

[@CrunchEconometrix]

I agree with your academic advisor. Thanks for the positive feedback. Deeply appreciated! 😊

[@imabou1395]

Please, What is the appropriate sample size in econonmetric studies? note I have panal data

[@CrunchEconometrix]

Hi Ima, observations must be > or = 30. For panel data that will depend on your research objectives whether to have a short panel (N > T) or a long panel (N < T).

[@imabou1395]

@@CrunchEconometrix thank you very much prof💗

[@biagiomeli3]

really thanks. i just wat to know if possible, why after the robust, the ANOVA table disappear?

[@CrunchEconometrix]

Hi Biagio, I cannot tell you why since it is Stata's configuration. Regards.

[@subhasreebanerjee8665]

If I use Hausman Taylor regression to estimate Fixed Effect panel data model (1) with heteroskedasticity, can I use this vce(robust) transformation? (2) Additionally, if there is serial correlation in the error term, how do we correct for that? Thanks!

[@CrunchEconometrix]

Subhasree, I have no videos on Hausman Taylor. You may want to check out other online resources. Thanks.

[@subhasreebanerjee8665]

@@CrunchEconometrix thanks!

[@MOHAMMEDHASSAN-gc3ed]

Nice being here!

[@CrunchEconometrix]

Thanks Mohd! Please may I know from where (location) you are reaching me?

[@MOHAMMEDHASSAN-gc3ed]

@@CrunchEconometrix Abuja, Nigeria.

[@mamadouniambele4176]

Dear madam. What about endogeneity ? May you suggest me an author to help me understand it ? Thank you so much madam.

[@CrunchEconometrix]

You may need to check other online resources, thanks.

[@tonimirotimi-ojo8421]

👏👏

[@CrunchEconometrix]

Thanks, Tonimi!

[@kidayantoko5422]

I would like to know whether is it possible to perform a panel Ardl in Stata 16?

[@CrunchEconometrix]

Hi Kidaya, I foresee no problem using Stata16 for the technique.

[@kidayantoko5422]

@@CrunchEconometrix Thank you Professor. I know that Stata 16 is having an issue to perform mg (mean group) in panel ardl. I have another issue. While running a ardl panel data, using the first difference(d.lun d.ldp......) the error term is positive but when I use the 2nd difference f the dependent variable (d2.lun d.ldp....) the error term turns negative which is consistent with the concept. How could you explain that?

[@CrunchEconometrix]

Hi Kidaya, that's not an issue. Go ahead and interpret your results. The ECT has its coefficient and so is the 2nd lag of the depvar.

[@kidayantoko5422]

@@CrunchEconometrix Thank you Professor

[@kidayantoko5422]

Hi Professor, I had estimated Pool-mean Group without a problem. I attempted the MEan-Group Model, but in vain. see below xtpmg d2.lnun d.lni d.lng d.lnp, ec(ec) lr(l.lnun lnim lng lnp) m > g replace invalid new variable name; variable name ec is in the list of predictors r(110); In order to perform the Haussman Test, I need to estimate the Mean-Group model, but Stata 16 doesn't allow it. Is it mandatory to peform the test? How can I overcome this issue? Many thanks

[@patriciatippy]

may i ask why the probability becomes 0?

[@CrunchEconometrix]

Hi Emmanuel, which probability are you referring to?

[@yuridornelas]

Is the command "robust" = gls?

[@CrunchEconometrix]

Not to my understanding though both control for heteroscedasticity.

[@jamiewilson9488]

Hi, I've been watching your video for a long time (almost a year) and it has helped me a lot learning techniques in running stata! Thank you so much..! I am now writing my final year dissertation and I have a question from one of your old videos. Here is the link; kzbin.info/www/bejne/rYO8lmOpbttghrs Q. I have two I(0) variables and three I(1) variables. And interestingly, I intended to see the correlation(or relationship) between each of the two I(0) variables and each of three I(1) variables, respectively. In other words, I planned to have 6 VAR equations (2x3) in the beginning. I did Johansen & EG tests for all of them, which says there is no cointegration relationship. However, in your video, you say I have to use Pesaran et al. Bounds test to see a cointegration relationship and then use the ARDL model. 1. Does it mean I, in my case, cannot use VAR models? if so, why..? I just do not know why it might not make sense. (My professor told me it would not cause any problem and I can go with VAR models.) 2. How do I use an ARDL model to see the two-way relationship between the variables?

[@CrunchEconometrix]

Hi Jamie, thanks for the positive feedback and kind remarks about my KZbin videos. Deeply appreciated! But your queries are too long and time consuming. Break them into bits and post them on the comment section of the respective KZbin video. Thanks.