Estimating the number of common factors in serially dependent approximate factor models

A simple data-dependent filtering method is proposed before applying the Bai-Ng method to estimate the number of common factors in the conventional approximate factor model. The asymptotic justification is provided and the finite-sample performance is examined.

► Bai-Ng factor number estimates can be biased upwards under serial dependence. ► We suggest filtering the panel with an AR(1) model before applying Bai-Ng criteria. ► The method is consistent under a broader range of serial dependence processes. ► The method performs well in small-sample Monte Carlo simulations.