Thanks you, i want to study impact of macroeconomic variables on stock returns .the number if firms is 35 and the study period IS 10 years with monthly observations.the macroéconomic variables remain fixed for all firms.what type if panel i will use thanks
In a study, if i conduct both fixed effect model regression (one independent variable turned out significant) and gmm regression (two significant variables turned out significant -including the FE model's variable). What should i include in the paper for panel data analysis? Or should i include both?
1) Kindly judge the following lines. According to the final regression mod: Fixed Effects model; ROA has negative impact on Npl ratio. However in GMM regression, Both the ROA and the CRAR has negative impact on NPL RATIO (dependent variable). 2) are there any preliminary tests to run or validate GMM regression?
@@aniksaha9925 The study has employed Hausamn test to determine whether FEM is appropriate or REM is appropriate. The results of Hausamn test indicates that FEM is more appropriate. (support with some references). (Interpret the results of FEM) For robustness the study has also employed GMM to estimate the regression coefficients. The results indicate that- ....
You can run GMM only under certain conditions. Please watch this video- th-cam.com/video/xuJwPuv4PKY/w-d-xo.html This video explain those specifications.
Thank you mam for the video on GMM. I'm trying to employ GMM on my data set and it is showing error in instrument variables. Could you please tell what is the reason ?
Ma'am to analyse the impact of mergers and acquisitions on leverge of a company which test can i use.. i have already used paired t test, regression for the first two objectives. Can you please tell me if GMM will go fine with it ma'am
Hi, I would like to test the interaction between the independent variables (FDI, trade, human capital, etc) in explaining the impact to the dependent variable (economic growth), is GMM method recommended? If so, I tried running my panel regression using the GMM method but it says 'number of instruments greater than number of observations'. I have 5 independent variables for the period 1993-2018. Looking forward to your reply, thank you.
One of the conditions for using the GMM method is that your number of cross section (N) must be greater than the number of years (T). Here, your T > N, which violates the basic condition.
Thank you so much for the informative video. I have a question, I am trying to run my panel regression using the GMM method you explained here, but it's giving me this error message "Number of instruments greater than number of observations." What could be the reason? I want to estimate an equation with 6 independent variables, with annual data for the period 1995 - 2021. Some data is missing though for some variables in certain years. Please help!
@@komalkanwarshekhawat_ Thanks so much. I tried the DOLS as illustrated here and it came out well. I am relieved. You are a lifesaver. You have earned a subscriber
Ma'am, if possible please guide how to perform interaction effect on STATA. i mean, do we directlymultipliy IDVs or is there any other way to include interaction. As i am beginner please guide 🙏
Look at results- If J statistics is close to 1 and probability J statistics is far from 1 then the regression model is a good fit. This is how you can judge your regression model if you have employed GMM Model.
Hi dear, thank very much for sharing I have panel data, the dependent variable is the real GDP growth rate, and I want to use the system GMM. As we know that GMM needs to use the lagged dependent variable. However, the GDP is in growth rate means that it's already lagged!! The question is: Can I use the first lag of real GDP growth, OR do I need to use the second lag of the dependent variable in GMM? Please guide me for the correct understanding Thank you
Maam how to get the AR(1) and AR(2) values for checking no serial correlation in eviews? I have seen all research papers report these two values in addition to hansen and sargan stats. please maam let me know
Hello mam... I'm a mba banking and finance student in uk.... I'm in my project stage. I selected the topic impact of digitalization... My supervisor ask me to do by panal data analysis by using eview... Actually I'm unaware of this.... Mam can you please help with this?
Refer this article - www.researchgate.net/publication/356170442_Evidence_of_a_Threshold_Size_for_Norwegian_Campsites_and_Its_Dynamic_Growth_Process_Implications-Does_Gibrat%27s_Law_Hold/figures?lo=1
Hello ma'am, Thank you so much, your video was very useful. I need to add dummy variables for regions to my model, but when I do that I get the error message "near singular matrix". Is it not possible to add categorical dummies to the GMM model or am I doing something wrong?
hello Mam I like the way you present things. I appreciate. but I need some precisions on the GMM model. when did we use the one-step GMM equation, two-step GMM equation, one-step, and two-step GMM system? also, there is any preliminary tests to run before applying GMM? in order words, what are the conditions to run GMM modles? Thanks to read and help me.
The conditions to run GMM model is already explained in this video. Kindly watch. I shall soon upload a video on difference between GMM one step, two step , dynamic etc. Stay updated .
hello mam I am trying to apply GMM in my study data set, I fail to understand the results, when The same data set is taken for POLS the model and variable come significant but when I am apply GMM, the variables become insignificant. can you please show me right way to do my analysis
The regression coefficients might be insignificant due to presence of any of the either or all- Multicollinearity, Heteroscedasticity, Non-normal distribution of data etc. First run these post- diagnostic tests as mentioned above 👆 and remove the presence of all if it is there. Then proceed with GMM. Hopefully this will help.
Dear Sir/Mam, I don't know how to interpret the results of J statistic and Prob(J-Statistic) in difference GMM? J statistic value for my analysis is 19.99 Prob(J Statistic) is 0.45. Kindly guide me how to interpret. Thank you
Look at results- If J statistics is close to 1 and probability J statistics is far from 1 then the regression model is a good fit. This is how you can judge your regression model if you have employed GMM Model.
Hi how can i resolve the problem of Number of instruments greater than number of observations using GMM Can you please show a practical example. Thanks
Look at results- If J statistics is close to 1 and probability J statistics is far from 1 then the regression model is a good fit. This is how you can judge your regression model if you have employed GMM Model.
your explanation is easy to understand thank you so much mam
Thank you so much for your kind words 😌
You are the best teacher mam........ 👍 I like your teaching style ❤️❤️
Thank you dear 🤗🤗
Your expertise in teaching has put many student's mind at ease...😀
thnku mam 💝
Grateful and thankfully overwhelmed 🤗😊
Well explained with adorable voice👌
Thank you Gagan 😊
Mam in instrumental variables we also select control variables with other IV's?
Thank You Madam for this beautiful video
Most welcome 🙏
Well explained 👏 🤗
Thank you dear ❤️
Good explanation
Thank you 🙏
Well explained mam
Thank you Navdeep 🤗🤗
well explained👌 keep it up mam
Thank you Prince 😊
Thanks you, i want to study impact of macroeconomic variables on stock returns .the number if firms is 35 and the study period IS 10 years with monthly observations.the macroéconomic variables remain fixed for all firms.what type if panel i will use thanks
You can perform - Fixed/ Random effects model, GMM, NARDL, FMOLS etc.
In a study, if i conduct both fixed effect model regression (one independent variable turned out significant) and gmm regression (two significant variables turned out significant -including the FE model's variable).
What should i include in the paper for panel data analysis?
Or should i include both?
You can report the results of both the models. First run FEM. Then for robustness run GMM and show its results.
1) Kindly judge the following lines.
According to the final regression mod: Fixed Effects model; ROA has negative impact on Npl ratio.
However in GMM regression, Both the ROA and the CRAR has negative impact on NPL RATIO (dependent variable).
2) are there any preliminary tests to run or validate GMM regression?
@@aniksaha9925 The study has employed Hausamn test to determine whether FEM is appropriate or REM is appropriate. The results of Hausamn test indicates that FEM is more appropriate. (support with some references). (Interpret the results of FEM)
For robustness the study has also employed GMM to estimate the regression coefficients.
The results indicate that- ....
Okay, hausman test came out as 0.000
Thank you so much for your help. Really appreciate it.
Hello Madam,
Do we need to check for stationarity of variables, before doing GMM
Thank You
Yes always.
Hi Madam, Thanks for your effort...Actually while applying the above a message of Instrument is greater than observations????
You can run GMM only under certain conditions.
Please watch this video- th-cam.com/video/xuJwPuv4PKY/w-d-xo.html
This video explain those specifications.
Thank you mam for the video on GMM. I'm trying to employ GMM on my data set and it is showing error in instrument variables.
Could you please tell what is the reason ?
Please check that your instrument variables should be less than the no. of cross-section units.
Ma'am to analyse the impact of mergers and acquisitions on leverge of a company which test can i use.. i have already used paired t test, regression for the first two objectives. Can you please tell me if GMM will go fine with it ma'am
Yes, you can run GMM. You can also run FEM/ REM.
Hi, I would like to test the interaction between the independent variables (FDI, trade, human capital, etc) in explaining the impact to the dependent variable (economic growth), is GMM method recommended? If so, I tried running my panel regression using the GMM method but it says 'number of instruments greater than number of observations'. I have 5 independent variables for the period 1993-2018. Looking forward to your reply, thank you.
How many cross- sections are there?
@@komalkanwarshekhawat_ 7 cross section
One of the conditions for using the GMM method is that your number of cross section (N) must be greater than the number of years (T). Here, your T > N, which violates the basic condition.
Please tell procedure for running system gmm
Or make video on this mentioning what to use when
Sure will soon upload a video on difference GMM , system GMM and the difference between the two.
Thank you so much for the informative video. I have a question, I am trying to run my panel regression using the GMM method you explained here, but it's giving me this error message "Number of instruments greater than number of observations." What could be the reason? I want to estimate an equation with 6 independent variables, with annual data for the period 1995 - 2021. Some data is missing though for some variables in certain years. Please help!
In this case, I would suggest run DOLS, FMOLS.
th-cam.com/video/tW-nx6qgyvA/w-d-xo.html
Watch this panel regression model .
@@komalkanwarshekhawat_ Thanks so much. I tried the DOLS as illustrated here and it came out well. I am relieved. You are a lifesaver. You have earned a subscriber
@@willardbhasa8327 Thanks! Glad to help 👍
Mam may I please know is it is possible to run Moderation or mediation in System GMM method of regression? please solve my doubt mam
Ma'am, Please guide how to perform interaction effect in Panel Data Analysis in Eviews.
Use STATA ....I would suggest
Ma'am, if possible please guide how to perform interaction effect on STATA. i mean, do we directlymultipliy IDVs or is there any other way to include interaction. As i am beginner please guide 🙏
@@nishant541 Multiply and then generate new variable.
Thanks a lot Ma'am 🙏
👌👌
🤗🤗
Mam,can you tell me the thumb rule of J stastics
Look at results- If J statistics is close to 1 and probability J statistics is far from 1 then the regression model is a good fit.
This is how you can judge your regression model if you have employed GMM Model.
Hi dear, thank very much for sharing
I have panel data, the dependent variable is the real GDP growth rate, and I want to use the system GMM.
As we know that GMM needs to use the lagged dependent variable. However, the GDP is in growth rate means that it's already lagged!!
The question is: Can I use the first lag of real GDP growth, OR do I need to use the second lag of the dependent variable in GMM?
Please guide me for the correct understanding
Thank you
Your Welcome .... Use first lag. That will do the job. 👍
Mam kindly upload a video on how to do System GMM method in eviews
Sure, I shall upload soon.
@@komalkanwarshekhawat_ thanks a lot mam
Maam how to get the AR(1) and AR(2) values for checking no serial correlation in eviews?
I have seen all research papers report these two values in addition to hansen and sargan stats. please maam let me know
Very Good and Clearly Explained. But please explain that it is J-statistic or Z-statistic, as it sounds Z-statistic, but results have J-statistic.
Thank you so much. It is J statistics.
Mam can u please post video on how to choose between system GMM and Difference GMM method.
Yes, sure. Shall upload soon.
Thank you mam
Hello mam...
I'm a mba banking and finance student in uk.... I'm in my project stage. I selected the topic impact of digitalization... My supervisor ask me to do by panal data analysis by using eview... Actually I'm unaware of this.... Mam can you please help with this?
See my video on - 'How to arrange panel data' first
Hi mam
Is this the first step difference gmm or two step difference gmm
Difference GMM One step
Can you examine long-run and short-run effects using IV-GMM?
hello! your video is super helpful, may i ask where can i add the control variables? thank you
Refer this article - www.researchgate.net/publication/356170442_Evidence_of_a_Threshold_Size_for_Norwegian_Campsites_and_Its_Dynamic_Growth_Process_Implications-Does_Gibrat%27s_Law_Hold/figures?lo=1
Mam LN means is it log form of variables?
Yes, dear.
@@komalkanwarshekhawat_ thank you mam
Which is the best J-statistics p-value range?
We can't say what is the best value of J statistics. You interpret the results based upon how far and near it is from the value zero.
@@komalkanwarshekhawat_ Okay, thanks.
Ma'am how to get the value of constant? It is always showing constant added in instrument list? Is there any way to get the constant value
In that case, You can run any other regression model.
thank you so much maam for your instant reply
Mam waiting for your video on choosing between system GMM and Difference GMM and it's interpretation. Please upload mam
Thank for constant prompting. Could upload yet due to other assignments. Shall create soon.
@@komalkanwarshekhawat_ thank you so much dear mam, will wait mam
How do one decide between difference GMM & system GMM and also one step and 2 step?
I shall soon upload a video on this.
@@komalkanwarshekhawat_ Also a video on system GMM. Thank you.
Hello ma'am, Thank you so much, your video was very useful. I need to add dummy variables for regions to my model, but when I do that I get the error message "near singular matrix". Is it not possible to add categorical dummies to the GMM model or am I doing something wrong?
hello Mam
I like the way you present things. I appreciate. but I need some precisions on the GMM model. when did we use the one-step GMM equation, two-step GMM equation, one-step, and two-step GMM system? also, there is any preliminary tests to run before applying GMM? in order words, what are the conditions to run GMM modles?
Thanks to read and help me.
The conditions to run GMM model is already explained in this video. Kindly watch.
I shall soon upload a video on difference between GMM one step, two step , dynamic etc. Stay updated .
My output is showing me NA (i.e my AR p-value). Why?
Could be due to Missing values.
hello mam
I am trying to apply GMM in my study data set, I fail to understand the results, when The same data set is taken for POLS the model and variable come significant but when I am apply GMM, the variables become insignificant. can you please show me right way to do my analysis
The regression coefficients might be insignificant due to presence of any of the either or all- Multicollinearity, Heteroscedasticity, Non-normal distribution of data etc.
First run these post- diagnostic tests as mentioned above 👆 and remove the presence of all if it is there.
Then proceed with GMM. Hopefully this will help.
Dear Sir/Mam,
I don't know how to interpret the results of J statistic and Prob(J-Statistic) in difference GMM?
J statistic value for my analysis is 19.99
Prob(J Statistic) is 0.45.
Kindly guide me how to interpret. Thank you
Look at results- If J statistics is close to 1 and probability J statistics is far from 1 then the regression model is a good fit.
This is how you can judge your regression model if you have employed GMM Model.
Mam error stating near singular matrix is shown. please guide what to do mam
This is a very common error that usually appears when variables are correlated.
@@komalkanwarshekhawat_ thank you mam
how to tell the significance between independent and dependent var.?
Sorry, I didn't get your question.
Hi how can i resolve the problem of
Number of instruments greater than number of observations using GMM
Can you please show a practical example. Thanks
Your variables in the data set should be less than the number of observations.
Hello, the model is very interesting can you share with me the Data ?
Please join my telegram channel. The link is in the description of the video. I shall share the data on telegram. Thanks.
interpret again the J-statistics and prob (J-statistics)
Look at results- If J statistics is close to 1 and probability J statistics is far from 1 then the regression model is a good fit.
This is how you can judge your regression model if you have employed GMM Model.
Mam waiting for your video on choosing between system GMM and Difference GMM and it's interpretation. Please upload mam
th-cam.com/play/PLLR4r6qT6b1hZEr48Ma7tcsThivWhHbyp.html
Second part will be uploaded soon.
@@komalkanwarshekhawat_ thank you so much dear mam