Greetings to all who can read this post, I would very much appreciate your input given the time constraints that I am experiencing with regards to my thesis.
I am conducting a panel data regression to test the validity of the EKC and comparing the experiences of high-income and low-middle income countries. The variable used are Production-based per capita carbon emissions as the independent variable. GDP as the independent. Additionally, a binary variable is used. Other control variables included are selected to account for the scale, composition, and technological effect of the EKC.
The hausman test was conducted and it revealed that the random effects panel regression is preferred. My questions are therefore as follows:
1) Do we need to test for Multicollinearity, Heteroskedasticity, Autocorrelation, and Normality following the use of the random effects model?
2) Which of the classical assumptions can be relaxed when using the random effects panel regression model?
3) Is there any tips/commands to remedy for classical assumptions that are not met?
Thank you for your help and input, it is very much appreciated.
I am conducting a panel data regression to test the validity of the EKC and comparing the experiences of high-income and low-middle income countries. The variable used are Production-based per capita carbon emissions as the independent variable. GDP as the independent. Additionally, a binary variable is used. Other control variables included are selected to account for the scale, composition, and technological effect of the EKC.
The hausman test was conducted and it revealed that the random effects panel regression is preferred. My questions are therefore as follows:
1) Do we need to test for Multicollinearity, Heteroskedasticity, Autocorrelation, and Normality following the use of the random effects model?
2) Which of the classical assumptions can be relaxed when using the random effects panel regression model?
3) Is there any tips/commands to remedy for classical assumptions that are not met?
Thank you for your help and input, it is very much appreciated.
