Thank you for taking us through this analysis. However, can you also take us through video on the linear and non linear analysis you used in your study. Thank you
Great suggestion! However, I have already included "Slope Homogeneity Test for Heterogeneous Panel Data Models in Stata", deployed in the same article. See link: th-cam.com/video/CTg8S_p3SFI/w-d-xo.html Moreover, a Modern Causality test (Dumitrescu-Hurlin Causality Test) is also used in this article and is explain in this video that I uploaded few weeks ago. See link: th-cam.com/video/NScQxvXuqpw/w-d-xo.html In upcoming videos, I'll include other tests employed in it as well, soon. Thanks.
Sir, thank you for your efforts. Very helpful. I want to ask some questions. I will be very grateful. 1- slopes have to be homogeneous or heterogeneous? 2- As the name of the model Augmented CSARDL, does dependent variable can be I(0), or it has to be I(1)? Thank you again.
Thank you for posting this video. There aren't many resources on CS-ARDL since it's such a new estimation technique, so this was really useful. One question: how do you decide what value should cr_lags take? In your analysis you went with 0, why?
Thank you for appreciation. I'll post more of contemporary (as well as classic) material in near future. This was a simple model for demonstration purpose. This command has many augments/switches, which can be read in its manual. Basically, lags of the cross-sectional averages are added to account for cross-sectional dependence. For a more meticulous analysis, one should determine the cross-sectional dependence at variable level. Then, the number of lags can be variable specific. The same order as in cr() applies, e.g. if cr(y x1 x2), then cr_lags(1 1 2). Chudik and Pesaran (2015) suggest that estimator gains consistency if the floor of pT=(T)^(1/3) lags of the cross-sectional averages are added. [T=Time dimension of Panel dataset] In addition to statistical analysis, theoretical reasoning can also be used for lag selection.
In the CS-ARDL estimation technique, the cr_lags option is used to specify the number of lags of the cross-sectional averages. The value of cr_lags can be chosen based on the specific characteristics of the data and the research question. The choice of cr_lags depends on the following factors: Data characteristics: If the data exhibits strong cross-sectional dependence, it might be necessary to include more lags of the cross-sectional averages to capture the dynamics of the data. Research question: The choice of cr_lags should be guided by the research question and the specific hypotheses being tested. If the research question focuses on the long-run relationship between variables, it might be necessary to include more lags of the cross-sectional averages. Model performance: The choice of cr_lags can also be influenced by the performance of the model. If the model does not converge or the results are not robust, it might be necessary to adjust the value of cr_lags to improve the model's performance. In summary, the value of cr_lags should be chosen based on the specific characteristics of the data, the research question, and the performance of the model. It is recommended to experiment with different values of cr_lags to find the optimal value for the specific analysis.
you should conduct some pre estimation tests on your data to check its health. However unlike some other models, the CS-ARDL approach does not require the explicit estimation of short-run coefficients, as it focuses on the estimation of long-run coefficients and the error correction term. Therefore, the absence of short-run coefficients in the CS-ARDL results is a characteristic of this approach and not necessarily an indication of an error or problem in the estimation process.
Aoa sir i am following your paper n work regarding cs-ardl and putting the same commands with my variables but it is not working. Stata shows different errors what would be the reason?
Sir how to solve if the results are not significant? I have tried with or without log of both dependent and independent variables but my results are not significant
Desirable results are not always the case. Methodology brings rigor from the statistical side. Model building, however, considerably depends on theoretical reasoning. So You may try experimenting with other proxies of the same variables or some other variables. At the end, it is a painful mixture of trial and error.
![[LIVE] : ONE ลุมพินี 80 | คู่เอก "รักษ์ vs ยอดนำชัย"](http://i.ytimg.com/vi/ckLMRorsMeY/mqdefault.jpg)
Thank you for taking us through this analysis. However, can you also take us through video on the linear and non linear analysis you used in your study. Thank you
Thank you so much for this. Was stuck carrying out an analysis, and your video was really helpful and ive made significant progress.
Regards.
It would be great if you could provide an example of the analysis in the article you showed. Thanks lot.
Great suggestion!
However, I have already included "Slope Homogeneity Test for Heterogeneous Panel Data Models in Stata", deployed in the same article. See link:
th-cam.com/video/CTg8S_p3SFI/w-d-xo.html
Moreover, a Modern Causality test (Dumitrescu-Hurlin Causality Test) is also used in this article and is explain in this video that I uploaded few weeks ago. See link:
th-cam.com/video/NScQxvXuqpw/w-d-xo.html
In upcoming videos, I'll include other tests employed in it as well, soon.
Thanks.
Done a great job ...Jazak Allah
Welcome
Sir, thank you for your efforts. Very helpful.
I want to ask some questions. I will be very grateful.
1- slopes have to be homogeneous or heterogeneous?
2- As the name of the model Augmented CSARDL, does dependent variable can be I(0), or it has to be I(1)?
Thank you again.
welcome
1. CSARDL allows slopes to be heterogeneous.
2. in CSARDL model, the dependent variable can be integrated at level I(0).
okay I will make some videos related to it
Thank you for posting this video. There aren't many resources on CS-ARDL since it's such a new estimation technique, so this was really useful. One question: how do you decide what value should cr_lags take? In your analysis you went with 0, why?
Thank you for appreciation. I'll post more of contemporary (as well as classic) material in near future.
This was a simple model for demonstration purpose.
This command has many augments/switches, which can be read in its manual.
Basically, lags of the cross-sectional averages are added to account for cross-sectional dependence.
For a more meticulous analysis, one should determine the cross-sectional dependence at variable level.
Then, the number of lags can be variable specific.
The same order as in cr() applies,
e.g. if cr(y x1 x2), then cr_lags(1 1 2).
Chudik and Pesaran (2015) suggest that estimator gains consistency if the floor of pT=(T)^(1/3) lags of the cross-sectional averages are added. [T=Time dimension of Panel dataset]
In addition to statistical analysis, theoretical reasoning can also be used for lag selection.
@@drbilalmehmood206 i have similar problem.. How can I determine optimal lag? is there any test for it?
In the CS-ARDL estimation technique, the cr_lags option is used to specify the number of lags of the cross-sectional averages. The value of cr_lags can be chosen based on the specific characteristics of the data and the research question.
The choice of cr_lags depends on the following factors:
Data characteristics: If the data exhibits strong cross-sectional dependence, it might be necessary to include more lags of the cross-sectional averages to capture the dynamics of the data.
Research question: The choice of cr_lags should be guided by the research question and the specific hypotheses being tested. If the research question focuses on the long-run relationship between variables, it might be necessary to include more lags of the cross-sectional averages.
Model performance: The choice of cr_lags can also be influenced by the performance of the model. If the model does not converge or the results are not robust, it might be necessary to adjust the value of cr_lags to improve the model's performance.
In summary, the value of cr_lags should be chosen based on the specific characteristics of the data, the research question, and the performance of the model. It is recommended to experiment with different values of cr_lags to find the optimal value for the specific analysis.
@@drbilalmehmood206 huge thanks
welcome and don't forget to share
Excellent sir mashallah
welcome
Please, do you have more videos on CS-ARDL? Please make more videos on it. thanks
This video considerably explains this technique. However, some related tests are yet to be uploaded. Yes, I'll upload them.
@@drbilalmehmood206 I will be happy to see it. great work, Dr
thanks for explaining ect
Glad to help
Very nice 👍 Sir G
Thanks,
Check this one.
Its related.
th-cam.com/video/CTg8S_p3SFI/w-d-xo.html
Dear Sir, i am only getting long run cofficients. Please tell me the problem.
you should conduct some pre estimation tests on your data to check its health.
However unlike some other models, the CS-ARDL approach does not require the explicit estimation of short-run coefficients, as it focuses on the estimation of long-run coefficients and the error correction term. Therefore, the absence of short-run coefficients in the CS-ARDL results is a characteristic of this approach and not necessarily an indication of an error or problem in the estimation process.
Aoa sir i am following your paper n work regarding cs-ardl and putting the same commands with my variables but it is not working. Stata shows different errors what would be the reason?
wasalam.
Garbage in Garbage out...remember this when analysing the data.
make sure your data is clean and free of errors.
Can u explain it on eviews....plz
as per my information, it's eviews application is not built-in todate
it's better to use stata till it gets built-in in eviews
Sir how to solve if the results are not significant? I have tried with or without log of both dependent and independent variables but my results are not significant
Desirable results are not always the case.
Methodology brings rigor from the statistical side.
Model building, however, considerably depends on theoretical reasoning.
So You may try experimenting with other proxies of the same variables or some other variables.
At the end, it is a painful mixture of trial and error.
Can we ask your whatsapp contact please
+92 335 4219555
The pvalue of the CD test is significant. What does it mean ?
it infers that CSD exists