Monte Carlo Estimation of Heterogeneity Effects in a Panel Data Regression Model

Abstract

Violation of homoscedasticity assumption in a Panel Data Regression Model (PDRM) implies unequal variability of error terms, and this creates heterogeneity problem in estimation. This research thus attempts to investigate this phenomenon by extending the works of Baltagi et al. (2010) and Adeboye and Agunbiade (2017) within the context of fitting an audit fee model via a simulated panel data and its estimation through the derivation of a Conditional Lagrange Multiplier (CLM) test for heteroscedasticity given zero first-order serial correlation via a two-way error component model. Monte Carlo simulations were carried out for 27 different variations, of which its design assumed a uniform distribution under a linear heteroscedasticity function. Each of the variation was iterated 1000 times and the assessments of estimators considered are based on Absolute bias (ABIAS) of parameters estimates. Nine (9) different models at different specified conditions were fitted, and the best-fitted model is that given by a within estimator when heteroscedasticity is severe. This study established that using CLM test; the results provide good size and power at 5% significance level for the specified linear form of heteroscedasticity function with α assigned values 0, 1 and 2 denoting homoscedastic individual specific error, moderate and severe heteroscedastic errors respectively.