Testing for no factor structures: On the use of Hausman-type statistics

Castagnetti et al. (2015) propose two max-type statistics to test for the presence of a factor structure in a large stationary panel data model. In this contribution, we study the use of Hausman-type statistics based on the CCE estimator of Pesaran (2006) and the IE estimator developed by Bai (2009; see also Song, 2013). We show that tests based on either estimators cannot be employed directly, and either need further assumptions than those required for the simple purpose of estimation, or bias corrections, or cannot be used altogether.