Thank You sir for the explanations., as you explained here I used all the three tests for cointegration...the Kao cointegration test failed to reject the null hypothesis of no cointegration but Pedroni and Westerlund rejected the null hypothesis of no cointegration.., is that possible to say there exists a long-run relationship between the dependent and independent varialbes
Hi Sir. Following the question below, if my variables are all I(1), I should conduct the cointegration test on variables before differencing it right? My result for Pedroni shows Modified Phillips-Perron t is 5%. Given these mix result, can I say all the variables are cointegrated?
The essence of cointegration is to ascertain if a long run relationship exist in the model despite the non-stationarity nature of the data. For data that are already stationary, there will be no need for cointegration test. So, conduct stationarity test first, if all the variables or some are not stationary, you can now proceed to cointegration test.
Hello, I would like to know how the Westerlund (2007) unit root test is done on stata version 16 ......... knowing that it includes 4 types of robustness test
Hello sir, thank you very much. If my dependent variable is I(1) and the residual three independent variables are both of order I(0) and I(1) in my panel dataset T>N, which test should I perform, and if these tests are only for variables integrated of the same order or can be used while working with the combination of both I(0) and I(1) variables? Thanks once again!
@@halil8270 Surprisingly, today I finished working on my master's thesis using panel ARDL. I estimated results for three alternative estimators MG, PMG, and DFE, and used Hausman (1978) test to confirm the most appropriate method.
Hi. There is no need to generate the series by differencing. Differencing itself shows they are short run variables. You could still conduct stationarity test on the I(1) series to see if there is still convergence in the long run. The essence of cointegration test is to ascertain whether or not there is a long run relationship in the model. It they are not cointegrated, only the short run part of the model is valid
Thanks so much for creating this video.
Thank you!!!
Thank You sir for the explanations., as you explained here I used all the three tests for cointegration...the Kao cointegration test failed to reject the null hypothesis of no cointegration but Pedroni and Westerlund rejected the null hypothesis of no cointegration.., is that possible to say there exists a long-run relationship between the dependent and independent varialbes
Hi Sir, thank you for the video.
Could l use Panel ARDL without applying cointegration test? Is Cointegration prerequisite before panel ARDL?
good day sir, in running for cointegration tests, are we going to use the level form or the stationarized data? thank you.
Also share the data of this video, it will be better to learn.
can you conduct another class on how to estimate PVAR in stata?
Alright. I ll look into it
Hi Sir. Following the question below, if my variables are all I(1), I should conduct the cointegration test on variables before differencing it right? My result for Pedroni shows Modified Phillips-Perron t is 5%. Given these mix result, can I say all the variables are cointegrated?
The essence of cointegration is to ascertain if a long run relationship exist in the model despite the non-stationarity nature of the data. For data that are already stationary, there will be no need for cointegration test. So, conduct stationarity test first, if all the variables or some are not stationary, you can now proceed to cointegration test.
Hello, I would like to know how the Westerlund (2007) unit root test is done on stata version 16 ......... knowing that it includes 4 types of robustness test
Hi Wilfred. I used 15 currently and do not have access to 16 yet. I ll look around and get back. Thank you
Hello sir, thank you very much. If my dependent variable is I(1) and the residual three independent variables are both of order I(0) and I(1) in my panel dataset T>N, which test should I perform, and if these tests are only for variables integrated of the same order or can be used while working with the combination of both I(0) and I(1) variables? Thanks once again!
@@halil8270 Surprisingly, today I finished working on my master's thesis using panel ARDL. I estimated results for three alternative estimators MG, PMG, and DFE, and used Hausman (1978) test to confirm the most appropriate method.
Im inputing the command but stata cannot recognise it
sir , if our series are stationary at I(1)...do we have generate the series by differencing and then apply cointegration???
sir plzz answer my query?
Hi. There is no need to generate the series by differencing. Differencing itself shows they are short run variables. You could still conduct stationarity test on the I(1) series to see if there is still convergence in the long run. The essence of cointegration test is to ascertain whether or not there is a long run relationship in the model. It they are not cointegrated, only the short run part of the model is valid