Data driven robust estimation methods for fixed effects panel data models

The panel data regression models have gained increasing attention in different areas of research including econometrics, environmental sciences, epidemiology, behavioural and social sciences. However, the presence outlying observations may often lead to biased inefficient estimates model parameters resulting unreliable inferences when least squares method is applied. We propose extensions M-estimation Exponential squared loss function-based approaches with a data-driven selection tuning achieve desirable level robustness against outliers without estimation efficiency. consistency asymptotic normality proposed estimators also been proved under some mild regularity conditions. finite-sample performance our methods examined via several Monte Carlo experiments their results are compared ones from existing methods. In addition, macroeconomic dataset analysed using demonstrate superiorities.