I noticed in the literature (and forums) that some people state that time-varying covariates aren't an issue and to treat them the same as time-invariant covariates in STATA while others state that between and within-person effects may be different and that you need to separate the two. I realize STATA can recognize time-varying for time-invariant covariates (level 1 and 2). From the standpoint of specifying the Stata syntax, do you need to do anything different with level 1 and level 2 variables?
One way of separating between and within-person effects is described in the following: https://www.statalist.org/forums/for...r-mixed-models
However, I am not sure if this is necessary if the research question involves quantifying the effect of certain factors on the slope of a repeatedly measured dependent variable (i.e. using mixed depvar i.predictor##c.time )
One way of separating between and within-person effects is described in the following: https://www.statalist.org/forums/for...r-mixed-models
However, I am not sure if this is necessary if the research question involves quantifying the effect of certain factors on the slope of a repeatedly measured dependent variable (i.e. using mixed depvar i.predictor##c.time )
