Is it possible to identify whether and how cross-sectional variation and/or time-series variation contributes to the coefficients estimated from a panel data regression (e.g., xtreg, fe or xtreg, re)? If so, which command can be used to detect this? Thanks!