Panel Data. Fixed effect and Random effect. Model Two. EVIEWS

[@lailulmarom1422]

Thank you so much sir, your slow language make me understand well all your words. Thank you so much more more

[@sayedhossain23]

+Lailul Marom 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]

+Lailul Marom You are welcome. Please join Hossain Academy Facebook to share your ideas with us.

[@tomasrehurek4883]

Hi, can you answers a question for me?.. in video 24:50,,you estimate (again) FE model (this is your words) but you didn´t go to the next part of this "window" I mean "Panel opt."...and my question is what should be in "Panel opt." on the upper rectangle Fixed or None. Thank you for your answer. Tom PS: Very good video! :)

[@areef788]

Hi. I already know that fixed effect model is better than random effect model by using Hausman test. But may I know how can I check whether pooled reg or fixed effect is appropriate since my data doesn't have any dummy variables? thanks.

[@tanglimchhoung9124]

Thank you very much for the video. By the way, i have some problems with random effect model testing. When i run, Eviews provides comments " Random effects estimation requires number of cross sections > number of coefficients for between estimator for estimate of RE Innovation variance". I do not know how to solve the problem besides dropping some variables. Thank

[@HAYUDB]

Could someone reply, how to generate the dummy variables? I haven't found any satisfying tutorial so far, please. I have 6 countries, 1990-2014 (25 yrs of observation), 1 dependent variable and 4 independent variables.

[@shamsunnahar9590]

hi if I write 3 line formula in my estimation equation box I am getting "near singular matrix' command. do you have any suggestion how to avoid this?

[@abebebelew2056]

Your presentation is good, but we need more on panel data analysis and other models.

[@eyasyama]

How can explain the coefficients of dummy variables you get in the fixed effect model

[@nguyenvuphan4553]

Thank you! This is dead on what I am looking for!

[@eyasyama]

Thanx prof , plz how can i read the coeffecients results for the dummy variables in your fixed affect model? and how can i design the dummy data variabels plz i need your help thanx prof

[@163darkknight]

Can you answers a question for me? I have data and I use fixed-effect model. I want to creat a dummy that equal 1 for some company and equal for the rest but when i run, it announced "Near singular matrix". Why?

[@sayedhossain23]

When your data setting is not appropriate, EVIEWS provide this comment

[@sudipdey8767]

Dear Sayed thanks to you for your helpful videos. I have an question can I run fixed effect and random effect model without dummy varibale according to hausman test????

[@mayureshg3209]

hello, can i use this panel data analysis for time series data on macroeconomic indicators ? kindly elaborate the meaning of dummy variable and what data shoulld be used? thanks mayureesh

[@musumeputhry628]

anyone knows how to use chow test if i had dummy variable in eviews?

[@neilsavy3234]

Hi When i attempted the hausman test in eviews, i got a prob value equal to 1 and below in a box in mentioned " Cross-section test variance is invalid. Hausman statistic set to zero." where do you think I have gone wrong?

[@sayedhossain23]

Neil Savy Please make question in Hossain Academy facebook group below. facebook.com/groups/hossainacademy/

[@TrungNguyen-ov4rc]

hi, Mr,Sayed Hossain, I used random effect for panel data, and I don't know how to conduct some tests for multi-colinear, autocorrelation, heteroskedasticity. Please, show me how to me do these tests.

[@sayedhossain23]

Yes those tests can be done by using STATA software. You can see some of my videos using STATA...

[@ahmedfayed1546]

Thank you Syed, I have one question, I have a set of panel data and I ran each of OLS, Fixed Effects and Random effects models, then I ran Wald test to check for OLS vs Fixed Effects and ran the Hausman test to check Fixed Effects vs Random Effects. In both results I found each of OLS and Random effects models as more appropriate than Fixed Effects model however I am still in dilemma on how to compare now which is more appropriate of OLS vs Random Effects, please advice what test I am still missing in order to check OLS vs Random Effects, thx again.

[@sayedhossain23]

Please see all my videos using EVIEWS for Panel Data where I have shown which model would be most appropriate. Your answer is there.

[@binhminhmua2012]

Thanks Sayed. My question is how to select lag for model with panel data

[@bachblues2]

Thanks for this very intersting explanation on how to apply panel data analysis. However, and for argument 's sake, the fact that demand for computers is positively related with prices might signify that computers are rather luxury goods, where price elasticity of demand is so low that demand does in fact increases in the same direction as prices. That the cost of repairs is not statistically significant reasserts this perception . People just buy laptops that last longer and its more practical to buy a new computer, rather than repairing it, unless its aout a minor maintenance nuances and its cost is negligible. .

[@eyasyama]

Thanx also prof i get the answer for dummy var omitted,,,, thank you very very much for ur lecturers ,,,

[@yusmitashafa9642]

Hi, I have question... Now, I'm doing research about the effect of NTM on cofffee export by using gravity model with panel data. I ran data, i.e value of export, real gdp, distance and population as well as dummy sps and tbt, by using eviews. Without using dummy, I got the result. But, when I process data with dummy i didn't get the result. eviews notified it with "near singular matrix. What does near singular matrix mean? What should i do? please respond my question asap? thank you

[@sayedhossain23]

Yusmita Shafa I have not developed gravity model yet but have plan in future but if you join our Hossain Academy Facebook group probably our members may help you. The link is below. facebook.com/groups/hossainacademy/

[@akshaygopaul6595]

Dear Sir, for my Hausman test i'm having HAusman Statistic error, chi square value of 0.0000.. Any suggestion

[@quangminhinh4009]

can you help me how to decide which model (Random effect model or Pooled regression model) is appropriate? Thank you

[@sayedhossain23]

+Quang Minh Đinh Hossain Academy has videos on all these issues using STATA and EVIEWS.

[@debbyamalia1305]

i have a question, what about choosing between pooled vs random effect ? can you explain to me please how to do it? is it using LM test? how to run the LM test on this case?

[@sayedhossain23]

It is model two. Did you see model one? In model one of panel data in Hossain Academy you will find other example. Hossain Academy is located at www.sayedhossain.com

[@seenmunchan6586]

Dear Sir, I would like to know do we need to full fill CLRM assumptions for every cross section equation in panel data? For example I have 4 countries with 20 years time series data that produce 80 observations in panel; do I need to produce BLUE estimators in 4 countries' equations before I pool them to panel data?

[@nourzeineldine787]

Dear Dr. Hossain, I have 5 banks, 13 periods/65 observations, I'm in order to find the relation between the stock price and four independent variables (Dividend per Share, Return on Assets, Traded Volume and Liquidity ratio). In order to apply the model two (that you published), I don't know how to create the dummy variables of my data, because model two starts directly with dummy variables and I don't know how you creates them. I need your help please to know how can I create my dummy variables. Thank you Dr.

[@nourzeineldine787]

Moreover, at the beginning you said that there is 5 dummy variables because we have 6 countries, and later, you said that we have 5 dummy variables because of five countries. I want to know how many dummy variable should I create in my study and how can I create them. With all my respect Dr. Hossain.

[@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/

[@GhulamMustafa-by1bk]

Sir how to perform difference and system GMM in Stata or eveiws. Is any video uploaded??

[@farshidfarhan6121]

Hello Dr Hossain, it's so nice to share my problem, i have a problem with running of my data, there are 4 countries, 15 periods/ 60 observations. therefore when i run my model at first time by random effect the results are significant and positive but after testing hausman test, it prefer fixed effect but the fixed effect's result is not significant mean that the p value more the 5% . thanks i m looking for answer :)

[@sayedhossain23]

Farshid Farhan Hello Sir. Please join my Hossain Academy Facebook group. There we are discussing about models. Join and make question below link. Thanksfacebook.com/groups/hossainacademy/

[@sayedhossain23]

For details please check Basic Econometrics by Gujarati, 5th edition. You can also see the data file the way I have set dummy variables.

[@أحمدالشافعي-ط4ر]

I cannot find the data in your website, could you upload it again please or put it in the description box ... thanks

[@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/

[@MrSunArise]

thank you very much for the lesson. here i have a question. Which version of Eviews are you using? Why dont i have the option to chooose random effect model? And i dont have the option to run the Hausman test.. Can you please help ? thank you very much appreciate

[@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/

[@qaiserabbas2428]

Sir if sample size is 200 and the result of Hausman test is in favour of fixed effect model, so how many dummies we have to introduce in order to test the appropriateness of fixed effect model because you have said in this lecture that dummies will be created according to the number of companies

[@shamsunnahar9590]

what if the wald test probability is .0011 and .0007. which one is acceptable null or alternative?

[@sayedhossain23]

Both are less than 5 percent, so reject null hypothesis

[@shamsunnahar9590]

I was wondering as you said we need to check whether its zero. thanks very much

[@TableRocker7]

Hello Sayed, I have noted also your VAR videos here (thank you so much for this) and I was wondering if you have seen an application of VAR models in a panel data fashion such as to control for fixed or random effects. I am thinking of two time series, assume gdp growth and money supply for a sample of five countries. Could you perhaps provide a short example for eviews or stata? Many thanks in advance and please keep on sharing your videos.

[@sayedhossain23]

VAR panel I have not exploited yet, so unable to comment...

[@AmeenKhan-ls6mw]

how to manage control variable in our regression test in eviews??

[@sayedhossain23]

Dear Khan, 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/

[@evelynzheng8219]

Thank you so much! there has a question raised when i used the fixed effect model. Eviews shows "near singular matrixs" Is this mean some problems exist in my data?How to deal with it? Thank you, looking for your reply

[@psvthabest789]

+EVELYN ZHENG I have got the same problem!

[@jihaneaayale5280]

When there are dummies in all observations , the constant· term· has to be· excluded . If a constant· term· is included in the regression , it is important· to exclude· one of the dummy variables from the regression , making this the base· category· against which the others are assessed . If all the dummy variables are included , their sum· is equal to 1 (which stands for the variable X· 0· to the constant· term· B 0· ) , resulting in perfect· multicollinearity . This is referred to as the dummy variable trap·

[@giadam9231]

Hello Mr Hossain, I would like to ask you about a problem I have encountered. I have run a Fixed effects model and Pooled OLS following your guidance on youtube and managed fine. However when I try to run a Random effects model I get the following error message: "Random Effects estimation requires number of cross sections > number of coefs for between estimator for estimate of RE innovation variance." My analysis consists of 4 countries, with each a dependant variable and 11 independent variables over a time period of 21 years. With no dummies used. I hope you can help me, Thank you

[@sayedhossain23]

Please join Hossain Academy Facebook below so that we can share our views there. Thank you Sayed Hossainfacebook.com/groups/hossainacademy/

[@sayedhossain23]

Probably there is a syntax error. Keep on trying. Thank you Sayed Hossain from Hossain Academy at www.sayedhossain.com

[@LossinDulhey]

Can someone tell me if it is necessary for the data to be stationary to do the panel data analysis?

[@sayedhossain23]

Thank you Lossin, 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/

[@joeytmy]

hi.. why my fixed effect model with dummy output is same with my pooled regression output?

[@sayedhossain23]

There are few ways to estimate fixed effect model. One way is by appointing dummy variables. So you are likely to get the same result.

[@joeytmy]

Sayed Hossain Thanks for advice

[@LondonPhD]

Eyas, he ran the regression model correctly. one dummy is left for reference. you may run into a dummy variable trap if you assign dummies for all six countries. just google "dummy variable trap" you will get an answer to your question

[@ajakmichael6060]

Hello everyone! I am a bit confused with how do I go about test robustness check in panel data

[@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/

[@reesespeanut4778]

You are excellent!! Thank so much!!!

[@gogielannanthakumar2741]

Can panel data be used for cointergration as well. Please advise thank you.

[@sayedhossain23]

Yes it is called panel cointegration.

[@chandralekha1231]

Sir what are these dummy variables in your study? just random numbers?

[@sayedhossain23]

No Eview does not put dummy automatically. You have to create dummy variables.

[@woOmGow]

Dear Mr Sayed Hossain i have problem my project is about evaluating mutual funds for Jan 2003 - Jul 2014 my total funds are 62 and not all of them start in same time they start in different months or year i have to sum all net asset value for all funds in one column and i have to find the return rate i use formula R= (net 2 -net 1) /net 1*100 or R= Ln (net2/net1)*100 after that i find the average by R/n n refer to number of funds at each time ....my question is... is my way to sum the net true or not ? i appreciate your help thank you

[@sayedhossain23]

Please read the Basic Econometrics by Gujarati, 5th edition where he explained the meaning of dummy variables in details.

[@ratihkertawardhani9443]

Sir, is it okay to add lagged independent variable in fixed effect/random effect regression? my equation is Yit=cit+Xit(-1)+Zit+eit

[@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.

[@alipaf2002]

Hi, sir do you have Panel VECM video please?

[@sayedhossain23]

+alipaf2002 Yes Panel VECM is available in EVIEWS section of Hossain Academy.

[@ForsellOfficial]

Hello! I am currently doing my bachelor thesis in Economics and I have a question. We are looking at how GDP per capita is influenced by shock-resilience and shock vulnerability between 2007-2010 for 28 countries. We are going to use a Fixed effects model for this. What type of model would you suggest for this?

[@sayedhossain23]

ForsellOfficial For this one you can run panel DOLS and panel FMOLS model. You can see videos on these two models in Hossain Academy's EVIEWS section at www.sayedhossain.com

[@map1388]

hello and tq for your great explanation, my question is about panel option tab, i have eviews 8 and 6 but I could not find it, could you please how can i activate this tab in estimation sheet? TQ

[@BahasDuitTanpaRumit]

Salam Sayed, May Allah reward your sharing knowledge. Currently I'm struggling learning how to analyze panel data using Generelized Method of Moment. Could you kindly give us a lecture about it? Jazallah

[@alexislixie516]

HI Mr Hossain, I got a question about why in this episode you use dummy variables here? I can't understand. Because in 'Panel Data. Fixed effect and Random effect. Model One', you didn't use dummy variables. I am confused. Can you see my question? Thank you so much~

[@amanpasha8990]

Professor Hossain, my model has multiple dependent variables and a single independent variable. How can I run regression on such model? it is a panel data

[@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/

[@KIENTHUCKINHTE]

Thanks so much. It helped me very much

[@sayedhossain23]

You are welcome

[@kudakwashejisinau]

Dear Mr Hossain may you kindly assist on the question. suppose l have run the regressions on Random Effect, Fixed Effect and Pooled OLS and l have perform the Hausman Test on which l accept the Null hypothesis that RE model is appropriate. suppose also l run the Wald Test and l reject Null hypothesis and accept Alternative hypothesis on which Fixed effect is appropriate. How will l choose the most appropriate Model between Fixed effect and Random Effect based on those situations (above).. can you shed more light on that situation. Best Regards,

[@sayedhossain23]

Dear Jisinau, 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/

[@deeg1385]

hes kuda

[@eyasyama]

Thanx London for ur contribution

[@danielosborn2721]

Hello, thank you for the video, it was extremely useful! Would you be able to make on on GMM/ 2SLS? It would extremely useful and there aren't any others on the internet.

[@arubawx]

Doesn't eviews put the dummies automatically? when asked to do the fixed panel in equations estimation? Or am I wrong?

[@ergeskelo]

I get this message: log of non positive number. what does it mean and how can I avoid it

[@sayedhossain23]

You can not make non positive number into log.

[@rismastg9712]

Sayed Hossain I need to make price data into return (for CAPM), but when I make it into return using LOG, I got the message "log of non positive number". Then how do I fix this? thankyou

[@sayedhossain23]

You can not convert the negative variable into log form, that is why it is coming like that. Thank you Sayed Hossain from Hossain Academy at www.sayedhossain.com

[@sayedhossain23]

You see the data setting of dummy variables. From there u can learn. For reading, Basic Econometrics by Gujarati, 5th edition.

[@aregawhailu7498]

please send also time series data

[@Zeeshanali92]

Marvellous effort Mr Hossain. can u tell that how to add time dummy variable in a similar sort of model.

[@sayedhossain23]

Dear Ali, 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/

[@dascaluiulia8145]

Hello, thank you very much for your video. It is really helpful. Only one thing is not clear for me: how do I know where to put 0 and where to put 1 for dummy variables? Thank you!

[@sayedhossain23]

Dascalu Iulia 1 refers to series of years and 0 refer to another series of years.

[@185283]

What if GDP, EXPORT, and IMPORT are not stationary? Does it matter?

[@sayedhossain23]

Dear Jung,, 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/

[@bachblues2]

Dr. Hossain.In the beginning of this presentation you specified five dummy variables as there were six countries. Then why are you inputting six dummy variables in the estimating equation ? In fact the output results assigns a very high probability value tothe last coefficient.

[@海顾]

Douglas Ungredda I think he use 5.. d2-d6

[@rezaprasetyo8506]

halo pak .. maaf mau tanya untuk data panel pakai lag 1 tahun gimana ya pak? tahapan nya di eviews .. terima kasih

[@sayedhossain23]

Reza Mahapatih Can not understand the language

[@elhampourmokhtar4505]

excellent. thanks so much

[@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/

[@linjoki]

HOW DID YOU ARRIVE AT 6 CROSS-SECTIONS?

[@sayedhossain23]

Dear Lina, 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/

[@eyasyama]

Q2 ,,, in ur example you have six countries but you use five dummy variables d1 d2 ......d5 , how come ,, thanx prof

[@ifeanyikingsley4203]

Am exciting

[@joshuabilon7559]

Bakit po kayo may dummy variables? (in english: why dummy variable is there?)

[@joshuabilon7559]

sana mapili

[@marinyoyankee]

Thank you very much

[@eyasyama]

I think it should be five countries not six ,,, i think there was mistake

[@shamsunnahar9590]

in this vedio you ur sample is from 6 countries however, your dummy is for 5 countries. why?

[@sayedhossain23]

Hi London Phd, Thanks for answer.

[@sayedhossain23]

My version is 7

[@ryanturner35]

It would be good if this was in English

[@sayedhossain23]

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

[@teshaioketch8080]

who is still here in 2019?

[@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/

[@faizahlee861]

thanxxxxxxxxxxxxxxxxxxxxxxxx alotttttt coud you show us granger couslty test var in panel data plzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz

[@sayedhossain23]

+Faizah Lee I have not done it yet

[@sayedhossain23]

+Faizah Lee Thank you. I would like to invite you to join “Analysis” in Facebook below to discuss about econometrics, statistics and data analysis using EVIEWS, STATA, R, SPSS, Minitab, Microfit, Lingo, Excel etc. Thank you, Sayed from Analysis Group. facebook.com/groups/449520105253087/

[@faizahlee861]

i would love too

[@faizahlee861]

+Sayed Hossain thnx alot sir am having study" effect debt euro erea of growth gdp panel data "

[@irvandinudianto3266]

you are very helpful but you talk very slow

[@sayedhossain23]

Dear Irvandi, 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/

[@sayedhossain23]

Hello Testing

[@najmehizadi9331]

Dear Mr.Hossain I have some questions about your Eviews teaching. How can I connect with you. Thank you

[@sayedhossain23]

You can write here

[@najmehizadi9331]

My question is about estimating ARDL model in Eviews. How can I write the model when I have more than one independent variable? And how can I remedy the serial correlation? Thank you

[@sayedhossain23]

Remedial of serial correlation is a difficult part..You have to go for trial and error process...If failed, go for First differenced Method prescribed in Gujarati book...I have also one video on First Differenced Method... You can take as many as independnet variables..see my ARDL model in Hossain Academy

[@cheesydingdong6435]

i-inne the benining

[@anny23108]

God man! no efficiency at all!