Variance Estimation for Cumulative and Longitudinal Poverty Indicators from Panel Data at Regional Level

This chapter addresses the issue of estimating sampling error of cumulative and longitudinal poverty indicators from panel data. It commences with a discussion on the differences between cumulative and longitudinal measures of poverty and inequality. The chapter introduces the most commonly used methods for cross-sectional variance estimation, with particular focus on the Jackknife Repeated Replication (JRR) method. It provides an extension of JRR to cumulative and longitudinal poverty and inequality measures. Next, the chapter presents some practical aspects of variance estimation, namely specification of sample structure variables and design effects and correlation. A description of differences between cumulative measures and measures of net change, while the special case of cumulating 3-years’ averages, with an empirical application, follows. The numerical illustrations presented are based on EU Statistics on Income and Living Conditions (EU-SILC) data from Austria and Spain. © 2016 John Wiley & Sons, Ltd.