Variance Estimation Using Poststratified Complex Sample

  • Kim, Kyu-Seong (Department of Computer Science and Statistics the university of Seoul)
  • Published : 1999.04.01

Abstract

Estimators for domains and approximate estimators of their variance are derived using post-stratified complex sample. Furthermore we propose an adjusted variance estimator of a domain mean in case of considering the post-stratified complex sample as simple random sample. A simulation study based on the data of Farm Household Economy Survey is presented to compare variance estimators numerically. From the study we showed that our adjusted variance estimator compensate for the under-estimation problem considerably.

Keywords

References

  1. Survey Methodology v.19 Conditional properties of post-stratified estimators under normal theory Casady, R.J.;Valliant, R.
  2. Sampling Techniques (3rd ed.) Cochran, W.G.
  3. Journal of Official Statistics v.13 Effects of post-stratification on the estimates of the Finnish labor force survey Djerf, K.
  4. Survey Methodology v.22 Stability measures for variance component estimators under a stratified multistage design Eltinge, J.L.;Jang, D.S.
  5. Survey Methodology v.23 Poststratification into many categories using hierarchical logistic regression Gelman, A.;Little, T.C.
  6. Journal of Royall Statistical Society. Ser. A v.142 Post stratification Holt, D.;Smith, T.M.F.
  7. International Statistical Review v.54 Post-stratification against bias in sampling Jager, P.
  8. Journal of Royal Statistical Society v.36 Inference from complex samples Kish, L.;Frankel, M.R.
  9. Journal of the American Statistical Association v.88 Post-stratification : A Modelers perspective Little, R.J.A.
  10. Model assisted survey sampling Sarnsal, C.E.;Swensson, B.;Wretman, J.
  11. Journal of the American Statistical Association v.82 Conditional properties of some estimators in stratified sampling Valliant, R.
  12. Journal of the American Statistical Association v.88 Poststratification and conditional variance estimation Valliant, R.