A method for evaluating the rank condition for CCE estimators

This paper proposes a binary classifier to evaluate the rank condition (RC) that is required for consistency of the Common Correlated Effects (CCE) estimator. The RC postulates that the number of unobserved factors, m, is not larger than the rank of the unobserved matrix of average factor loadings, \rho. The key insight in this paper is that \rho can be consistently estimated with existing techniques through the matrix of cross-sectional averages of the data. Similarly, m can be estimated consistently from the data using existing methods. A binary classifier, constructed by comparing estimates of m and \rho, correctly determines whether the RC is satisfied or not as (N,T) -> infinity. We illustrate the practical relevance of testing the RC by studying the effect of the Dodd-Frank Act on bank profitability.