A new stationarity test for heterogenous panel data with large cross-sectional dimension is developed and used to examine the growth rates of unit labour cost for 51 US states. The test allows for persistent cross-unit dependence in the form of unbounded long-run correlation matrices, for which a simple parameterization is proposed. A KPSS type distribution results asymptotically, if T and N converge to infinity sequentially (T first, followed by N). Some evidence against stationarity (short memory) is found for the examined series.

Redner/-in

Adina Tarcolea

Johann Wolfgang Goethe-Universität Frankfurt

Termin

19.07.2007 | 16:00-17:00

Veranstaltungsort

ZEW, L 7,1 D-68161 Mannheim

Raum

Hamburg

Kontakt

Kommissarische Leitung