The fixed effects model wastes a lot of degrees of freedom by estimating each individual effect. Since the individual effect is a set of time-invariant observable and unobservable influences, we will often not be interested in its coefficient. These estimates are not even consistent if sample grows large and the time span is fixed. Incidental parameter problem also arises in the traditional estimating approach. In this lecture we will learn how to estimate panel-data models with two or more types of fixed or random effects.