This Matlab code computes the panel unit root test statistics proposed by Moon and Perron (2002) for panels with dynamic factors. The user can put his (T,N) matrix of data (where T is the time dimension and N is the cross section one) and choose the deterministic component: with individual effects but no time trends or with individual effects and time trends. The user can also choose the kernel function and the selection method for the bandwidth parameter. The optimal number of common factors will be automatically determined according the chosen information criteria (see Bai and Ng, 2002). The code displays both the ta_bar and tb_bar statistics of the null hypothesis of a unit root and their p-value. For more detail see the survey of Hurlin and Mignon (2007).
International Journal of Research in Marketing (2019)
Gaustad Tarje, M. Bendik, Warlop Luk, and Fitzsimons Gavan J.