A panel data regression model in which it is assumed that unobserved heterogeneity is either constant over time for every cross-sectional unit (group-specific fixed effects) or constant across the units at every time period (time-specific fixed effects). This unobserved heterogeneity can be removed by subtracting the mean (over time in the former and across units in the latter case) from each observation, or by introducing binary variables indicating the cross-sectional unit in the former or time period in the latter case. See also between-groups estimator; random effects.