1 / 15

Imputation in EU-SILC survey

Imputation in EU-SILC survey. Andris Fisenko and Vita Kozirkova Central Statistical Bureau of Latvia. EU-SILC. "EU Statistics on Income and Living Conditions (EU-SILC)". This survey is conducted in all countries of the European Union. Variables. Total housing cost

laddie
Download Presentation

Imputation in EU-SILC survey

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Imputation in EU-SILC survey Andris Fisenko and Vita KozirkovaCentral Statistical Bureau of Latvia

  2. EU-SILC "EU Statistics on Income and Living Conditions (EU-SILC)". This survey is conducted in all countries of the European Union.

  3. Variables • Total housing cost • Family/Children related allowances • Social exclusion • Housing allowances • Income received by people aged under 16 • Regular taxes on wealth • Regular inter-household cash transfer paid • Regular inter-household cash transfer received • Other

  4. Structure • Housholds level • Income per year is asked. • If refusal, then ask average and how often. • If answer is “ I don’t know” then IMPUTATION • Personal level • Differences for persons level • Average in one time or interval.

  5. Imputation methods • RMV (Replace Missing Values) • Hot deck & Cold deck • Mean/Median imputation • Nearest Neighbour • Regression • Other

  6. SPSS RMV- 1

  7. RMV - results

  8. Cold-deck • Does not bring new information • Very good benchmarking method • There is not data from the past • Method will be analysed in the future

  9. Random donor (Hot-deck) • The donor is chosen randomly within same group • Random overall imputation is not sensible • But random donor within imputation classes (or clusters, groups, ..) is important method

  10. Random donor (Hot-deck) • Firstly, the data is divided into homogenous sub groups by a a suitable method, e.g. easily sorting into categories by some explanatory variable. Secondly, the donors are chosen randomly within these new groups. • Many clustering methods are using the idea of nearest neighbours when determining a new grouping of the data.

  11. Multiple imputation • Creates m imputed data sets. • Hot deck+Multiple imputation

  12. Some results

  13. Some results

  14. What's next • Software • Linear regression • Auxiliary information from • State Revenue Service • State Social Insurance Agency • Regional governments • Banks

  15. Thank you for your attention!!!

More Related