Hello all,
I am writing my Masters dissertation researching the effects of household debt on the wage share (share of national income distributed in wages) having compiled a panel dataset with 15 countries on a 30 year time period.
I am confused as to whether I have committed any grave statistical errors in setting up my model because the scatterplot shows a clear negative relationship (as hypothesized, see attached image,) while the variable in the time and country fixed effect regression shows a positive coefficient.
I am aware that this is a totally possible result considering that I am controlling the relationship with 5 other variables. Nevertheless, are there any tips on how I get check the robustness of my model or inquire as to what the reason for the change of coefficient are?
Many thanks in advance.
I am writing my Masters dissertation researching the effects of household debt on the wage share (share of national income distributed in wages) having compiled a panel dataset with 15 countries on a 30 year time period.
I am confused as to whether I have committed any grave statistical errors in setting up my model because the scatterplot shows a clear negative relationship (as hypothesized, see attached image,) while the variable in the time and country fixed effect regression shows a positive coefficient.
I am aware that this is a totally possible result considering that I am controlling the relationship with 5 other variables. Nevertheless, are there any tips on how I get check the robustness of my model or inquire as to what the reason for the change of coefficient are?
Many thanks in advance.
