PANEL FULLY-MODIFIED OLS(FMOLS), Dynamic OLS(DOLS), Canonical Cointegrating Regression(CCR) STATA 17

[@miqbal1405]

"Exceptionally informative"

[@drbilalmehmood206]

Glad it helped.

[@uzmaabubakar3239]

Informative video

[@drbilalmehmood206]

Glad you think so!

[@emmanuelakuoko-konadu6469]

Thank you

[@sarahahmedchawsheen5455]

Thanks for this useful video, can I apply either of FMOLS, DOLS, and CCR on heterogenous cross dependence panel data with T>N ?

[@drbilalmehmood206]

Pedroni, P. (2001). Fully modified OLS for heterogeneous cointegrated panels. In Nonstationary panels, panel cointegration, and dynamic panels. suggests that dynamic panels in a manner that is consistent with the degree of cross sectional heterogeneity. However, my humble opinion is that you should use CS-ARDL as the main technique, while using FMOLS (and DOLS, CCR) as the robustness test estimators.

[@sarahahmedchawsheen5455]

@@drbilalmehmood206 thank you so much.

[@sarahahmedchawsheen5455]

@@drbilalmehmood206 Dr. excuse me for too many questions. since I have cross section dependence of my panel with T>N should I use Pedroni Cointegration test or Westerlund cointegration test. ? because I did Pedroni test my data are not cointegrated while concerning Westerlund test my data is cointegrated. which one I have to adopt?

[@drbilalmehmood206]

It is possible for two tests to disagree even if both tests' assumptions are met. Pedroni's test only allows for very simple forms of cross-sectional dependence. So the rejection of that test might be spurious because not all underlying assumptions for the test to be valid are met. It is better to rely on Westerlund's test.

[@sarahahmedchawsheen5455]

@@drbilalmehmood206 thank you so much doctor Bilal.

[@aryaasvadishirejini8184]

Thank you professor mehmood for this useful content. I have some questions that l would like to ask you. I wanted to run a Feasible generalized least squared, (FGLS) model regression and my dependent variable is stationary at I(0) and the rest of independent variables are stationary at I(1). Also, my variables are cointegrated and the number of time periods are bigger than the number of cross-sections. Plus, my model is heteroskedastic and has autocorrelation and cross-sectional dependence. In your opinion, FMOLS or DOLS are the best research methods for my analysis? Or I'd better to find another method? If so, what would you recommend?

[@drbilalmehmood206]

Both FMOLS (Fully Modified Ordinary Least Squares) and DOLS (Dynamic Ordinary Least Squares) are suitable for your analysis. FMOLS and DOLS are methods used to estimate cointegrating regressions in the presence of endogeneity, autocorrelation, and heteroskedasticity. They are particularly useful when the variables are cointegrated and some of them are I(0) and others are I(1). FMOLS is suitable when the model is homoskedastic, while DOLS is appropriate when the model is heteroskedastic and has autocorrelation and cross-sectional dependence. Both methods are commonly used in empirical research to address the issues you have described. The choice between FMOLS and DOLS depends on the specific characteristics of your data and the assumptions of each method. It is recommended to compare the results obtained from both methods and consider the robustness of the estimates. Additionally, you may also consider other robust estimation techniques such as feasible generalized least squares (FGLS) to address the issues of heteroskedasticity and autocorrelation in your model. In summary, both FMOLS and DOLS are suitable for your analysis, and it is advisable to compare the results from both methods and consider the robustness of the estimates. Moreover, you may also explore other robust estimation techniques such as FGLS to address the specific issues in your model.

[@aryaasvadishirejini8184]

Thank you so much, Dr. Bilal Mehmood. Allah bless you and your family.@@drbilalmehmood206

[@drbilalmehmood206]

JazakAllah

[@Allen-qn2ru]

it doesn't show the t-stat for my results but beta, would you know why does it happen..? And thank you so much for this video, it help me so much

[@drbilalmehmood206]

Firstly, it happens with this command, sometimes. It can be due to the underline process of the estimation. So you may check the health of your data, by conducting some pre-estimation tests. Secondly, rather than getting t-statistics, the emphasis in DOLS and FMOLS regressions is frequently on estimating coefficients and evaluating their statistical significance. This is because serial correlation and endogeneity are frequently taken into account when adjusting the standard errors and t-statistics in these methods.

[@NikeeSilayachP20MS009]

Hello Sir, I have installed the command and when I am running the regression it is showing an error. ssc install xtcointreg checking xtcointreg consistency and verifying not already installed... installing into c:\ado\plus\... installation complete. . xtcointreg Y X1 X2 X3, est(dols) command cointreg is unrecognized r(199); Please help.

[@drbilalmehmood206]

hello please introduce yourself and then you may also share your data so that It could be checked.

[@gifmbewe537]

i faced the same problem. I am not sure whether you got help on this

[@toqeerabbas6634]

Sir which stata version supports this command Xtcointreg

[@drbilalmehmood206]

You should execute command, "xtcointreg". If it is not installed then, you can install it using following command. ssc install xtcointreg I have Stata 14.2 as well, and it supports this command. Which Stata version do you have and have you installed this command on it?

[@toqeerabbas6634]

@@drbilalmehmood206 thnxx sir I have 15 and 17 version But unfortunately after installing ssc install Xtcointreg Command not working

[@drbilalmehmood206]

does it get installed? or some error message? as you are writing Capital 'X' in xtcointreg. stata is Case sensitive and allows lowercase in commands keyword.

[@SegunBolarinwa-vk1ss]

I use stata 14 and unfortunately the xtcointreg is not working either@@drbilalmehmood206

[@drbilalmehmood206]

if possible, you can share your data and I can check this on my own system.

[@kinzanaseer4387]

Sir why this command xtcointreg is coming unrecognized

[@drbilalmehmood206]

because it is not installed on your stata. execute this command and it will be installed: ssc install xtcointreg

[@kinzanaseer4387]

@@drbilalmehmood206 I have installed it by using this command already but it's still coming unrecognized

[@drbilalmehmood206]

please inform about the nature of your work and then you may also share the data so that it could be checked.

[@gifmbewe537]

​@@drbilalmehmood206 how do i send you my data, I am facing the same problem

[@astitchintime5654]

Will it not work on Stata15??

[@drbilalmehmood206]

primarily this command should be installed so you should check if you have installed it or not. ssc install xtcointreg is the command to install check this and then let me know.

[@bushrakk445]

👍

[@drbilalmehmood206]

thanks

[@aymenyaqub1199]

👍

[@drbilalmehmood206]

Thanks