mereru-ka/TeaYield-Prediction-GA-XGBoost: coefficient analysis code

All empirical analyses in this study were conducted using Stata 17. The MMQR estimations were implemented with the mmqreg command across nine quantiles (0.10–0.90), while the cross-sectional time-series FGLS models were estimated using the xtgls procedure with correlated panels and AR(1) disturbances. In addition, linear panel regressions with panel-corrected standard errors (PCSEs) were performed via the xtpcse command. These complementary estimation strategies ensured robust inference by accounting for distributional heterogeneity, contemporaneous correlation, and serial dependence in the panel data structure.