Estimation of the variance of cross-sectional indicators for the SILC survey in Switzerland

View the most recent version.

Archived Content

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.

Archived

This page has been archived on the Web.

Anne Massiani¹

Abstract

SILC (Statistics on Income and Living Conditions) is an annual European survey that measures the population's income distribution, poverty and living conditions. It has been conducted in Switzerland since 2007, based on a four-panel rotation scheme that yields both cross-sectional and longitudinal estimates. This article examines the problem of estimating the variance of the cross-sectional poverty and social exclusion indicators selected by Eurostat. Our calculations take into account the non-linearity of the estimators, total non-response at different survey stages, indirect sampling and calibration. We adapt the method proposed by Lavallée (2002) for estimating variance in cases of non-response after weight sharing, and we obtain a variance estimator that is asymptotically unbiased and very easy to program.

Key Words

SILC survey; Rotating panel; Inequality indices; Variance estimation; Weight-share method.

Table of content

1 Introduction

2 Sample design and survey procedure

3 Cross-sectional household weighting

4 Linearization and approximation of variance

5 Variance estimation and weight sharing

6 Final formula

7 Numerical application and discussion

8 Conclusion

 

 

 

 

 

---

¹    Anne Massiani, Institute of Statistics, University of Neuchâtel, Pierre-à-Mazel 7, 2000 Neuchâtel, Switzerland & Federal Office of Statistics, Espace de l'Europe 10, 2010 Neuchâtel, Switzerland. E-mail: anne.massiani@unine.ch.