classical criteria, as well as forecast combination techniques with constant and non-constant weights. With respect to the … standard GARCH specification, the non-linear models generally lead to better forecasts in terms of both smaller forecast errors … and lower biases. In-sample forecast combination regressions are better than those from single Mincer …

. This could be the result of a bias caused by standard estimation error. Using German social security records we show that … the effect of this bias is substantial in samples with limited inter-firm movement. The correlation between worker and … ; limited mobility bias …