Variance estimation of change of poverty based upon the Turkish EU-SILC survey

Interpreting differences between point estimates at different waves may be misleading, if we do not take the sampling variation into account. It is therefore necessary to estimate the standard error of these differences in order to judge whether or not the observed differences are statistically significant. A major problem is to take into account of temporal correlations between estimators. Correlations play an important role in estimating the variance of a change between cross-sectional estimates. The standard correlation can be biased, because of the rotation of the design used for the European Union Statistics on Income and Living Conditions (EU-SILC) surveys. Furthermore, poverty rates depend on poverty thresholds which are estimated. We propose to use a multivariate linear regression approach to estimate the correlations. We also show how this approach can be adjusted to account for the estimation of poverty thresholds. The proposed estimator is not a model-based estimator, as this estimator is valid even if the model does not fit the data. We implemented the proposed approach to the Turkish EUSILC survey data