Abstract.Day2. Verbic

Statistical Matching of Different Microdata Sets

Matic Verbič, Matjaž Koman, and Tanja Istenič

Harmonized European NTA are estimated using income and consumption data from two separate micro-data surveys, the EU Statistics on Income and Living Conditions (EU-SILC) and the Household Budget Survey (HBS). in the past, the estimates from these two datasets were merged by simply imputing age- and sex-specific averages of private consumption in EU-SILC. However, for a more precise estimation of intra-household transfers, the newly estimated European NTA aim to explore different statistical matching techniques to merge EU-SILC and HBS data. Matching is based on common variables such as income, household type, tenure and other socio-economic characteristics and provides more detailed insight into consumption patterns in relation to income levels and other household characteristics. Techniques such as random hot-deck imputation and semi-parametric approaches are used for matching. We believe that this research can also help other national NTA teams attempting to estimate NTA using income and consumption data from different datasets.