Asymptotics for Non-Stationary Panels with High-Dimensional Nuisance Parameters

This paper studies asymptotic theory for a nonstationary panel autoregressive model when cross-sectional dimension (n) and time dimension (T) are large. We considers the nonstationary case in the presence of both cross-sectional and time fixed effects, which is not investigated in existing literature of dynamic panel models e.g. Hahn and Kuersteiner (2002) and Hahn and Moon (2006). We also provide the limiting distribution of the bias-corrected (quasi-) maximum likelihood estimator for both Gaussian and non-Gaussian error terms