Short T$$ T $$ dynamic panel data models with individual, time and interactive effects

This paper proposes a transformed quasi-maximum likelihood (TQML) estimator for short T $$ dynamic fixed effects panel data models allowing interactive through multifactor error structure. The proposed is robust to the heterogeneity of initial values and common unobserved effects, while at same time standard effects. It applicable both stationary unit root cases. order condition identification number established, conditions are derived under which parameters locally identified. shown that global in presence lagged dependent variable cannot be guaranteed. TQML proven consistent asymptotically normally distributed. A sequential multiple testing ratio procedure also estimation factors consistent. Finite sample results obtained from Monte Carlo simulations show determining performs very well, has small bias mean square (RMSE) correct empirical size most settings. practical use approach demonstrated by means two illustrations literature on cross county crime rates country growth regressions.