Dear Stata users,
Quick question. I have a strongly balanced panel data for 86 Russian regions for 2013 and 2018. The variables encompass: preventable mortality (dependent variable) as well as smoking habits, drinking habits, eating patterns, health systems variables, regional level of development, regional poverty levels (all independent variables). My idea was to use the variation of the variables across the regions and over time in order to help explain the variation in my dependent variable. Now, when I include a time dummy in the regression analysis, the significance in most of the independent variables disappears, while the year dummy is negative and significant. Thus, my question is: is the inclusion of time dummy in this type of panel set up desirable and necessary and, more importantly, what's the value added of adding the time dummy to the regression?
Grateful for any feedback that you might have!
Quick question. I have a strongly balanced panel data for 86 Russian regions for 2013 and 2018. The variables encompass: preventable mortality (dependent variable) as well as smoking habits, drinking habits, eating patterns, health systems variables, regional level of development, regional poverty levels (all independent variables). My idea was to use the variation of the variables across the regions and over time in order to help explain the variation in my dependent variable. Now, when I include a time dummy in the regression analysis, the significance in most of the independent variables disappears, while the year dummy is negative and significant. Thus, my question is: is the inclusion of time dummy in this type of panel set up desirable and necessary and, more importantly, what's the value added of adding the time dummy to the regression?
Grateful for any feedback that you might have!
