Generalized estimation equations method for fixed effects panel interval-valued data models

This paper studied panel interval-valued data models with individual fixed effects, in which the correlation within a group was considered and the group average method was used to eliminate the fixed effects. Then, we applied generalized estimation equations (GEEs) to analyze panel interval-valued data models and gave a computational algorithm to obtain the estimators. Some Monte Carlo simulations and real data analysis showed that, in contrast with the least-squares dummy-variable (LSDV) method, the proposed GEEs method has advantages in forecasting performance.