By exploring my dataset, I noticed great Variation among my Groups of the panelid. I thus want to know if this alone justifies to run a fixed effects model to Control for all the heterogeneity of my panelid or if a pooled model or random effects model with clustered Standard Errors, which tells Stata that the observations are correlated within the Group, will also account for the heterogeneity as much?