The paper provides information on sample sizes and panel attrition in the German Socio-Economic Panel (GSOEP) for the years 1984-1999. Furthermore, results of the sequential estimation of non-response probabilities for the different subsamples of the GSOEP are described in detail.

Using the panel data for the Philippines in 2003-2009, the paper estimates a three-level random coefficient model to measure household vulnerability and to decompose it into idiosyncratic and covariate components. It corrects heterogeneity bias using Bell and Jones's (2015) "within-between"...