EVALUATION OF PANEL DATA ESTIMATORS UNDER THE UNBALANCED PANEL DATA FOR SMALL DATA SIZES OCCASIONED BY MISSINGNESS
Description
A Simulated panel data from a sample of N (20 and 30) across time period of T = 4 and T= 6.
Files
Steps to reproduce
The data structure follows a normal distribution with ε_(1t )~N (0,1), where t= 1,…,T ; i = 1,…,n_k and k= 1,…,5. A total of k = 5 subjects (for instance) was studied and simulated over T = 4 and T= 6 times. Thus, a total of n = N (k×T) observations were generated. An independent variable x_(it )was considered and simulated from the Gaussian distribution; X_it~iid N(20,1). Over fixed time interval of k = 4 and 6, the vector of parameters, β is set as β ̂=(β ̂_0it,β ̂_1it )^/= (20,1)^/.The response variable Y_(it ) was simulated from a panel date regression model Y_(it )= β_0it+ β_1it x_1it+ ε_it. accordingly. An Unbalance time interval was infused into the data by randomly removing ω% of the total sample from the data, where ω take on values 5, 10, 15 and 20. (Balogun et al., 2022).
Institutions
- University of Ilorin