1: \begin{abstract}

2: 		In this paper we examine the existence of heterogeneity within a group, in

3: 		panels with latent grouping structure. The assumption of within group

4: 		homogeneity is prevalent in this literature, implying that the formation of

5: 		groups alleviates cross-sectional heterogeneity, regardless of the prior

6: 		knowledge of groups. While the latter hypothesis makes inference powerful, it

7: 		can be often restrictive. We allow for models with richer heterogeneity that

8: 		can be found both in the cross-section and within a group, without imposing

9: 		the simple assumption that all groups must be heterogeneous. We further

10: 		contribute to the method proposed by \cite{su2016identifying}, by showing that

11: 		the model parameters can be consistently estimated and the groups, while

12: 		unknown, can be identifiable in the presence of different types of

13: 		heterogeneity. Within the same framework we consider the validity of assuming

14: 		both cross-sectional and within group homogeneity,

15: 		using testing procedures. Simulations demonstrate good finite-sample

16: 		performance of the approach in both classification and estimation, while

17: 		empirical applications across several datasets provide evidence of multiple

18: 		clusters, as well as reject the hypothesis of within group homogeneity.

19: 		\newline\noindent{{Keywords}}: Panel Data Models, Latent

20: 		Grouping Structures, Heterogeneity tests. \newline

21: 		

22: 		\noindent{JEL Classifications}: C18, C23, C38, C51, C55.

23: 		

24: 	\end{abstract}

25: