Convergence rate of estimators of clustered panel models with misclassification

We study kmeans clustering estimation of panel data models with a latent group structure and N units T time periods under long asymptotics. show that the group-specific coefficients can be estimated at parametric NT-rate even if error variances diverge as T→∞ consequently some are asymptotically misclassified. This limit case approximates empirically relevant settings is not covered by existing asymptotic results.