A heteroskedasticity robust test for cross-sectional correlation in a fixed effects panel data model

Abstract This paper proposes a new test for detecting no cross-sectional correlation in a fixed effects panel data model. The newly proposed test is allowing for the heteroscedastic variances across time. It shows that if the error terms have zero skewness, the new test converges to a standard normal distribution as ( n , T ) → ∞ with n ∕ T → c ∈ ( 0 , ∞ ) ; if the skewness of the error terms is nonzero and finite, the test converges to a standard normal distribution as ( n , T ) → ∞ with n 2 ∕ T → 0 . Besides, the paper conducts Monte Carlo simulations for studying the finite sample properties, and the results verify our theoretical findings.