TY - CHAP
TI - Use of survey weights for the analysis of compositional data
T2 - Compositional Data Analysis, Theory and Applications, chap. 9
CY - Oxford
AU - Graf, M.
A3 - V. Pawlosky-Glahn
A3 - A. Buccianti
LA - en
PY - 2011
AB - This chapter contains sections titled:
- Introduction
- Elements of Survey Design
- Application to Compositional Data
- Discussion
- References
KW - use of survey weights - for compositional data analysis;surveys, built to optimize - estimation of totals in population subgroups, number of variables;key variable chosen, design optimized - for variable, between cost and precision;analysis of compositions, a byproduct - design, never optimized for compositions;survey design elements - data collection in finite population surveys, existence of registers;randomization - individual elements of population, and sampling frame;design-based estimation - estimation of statistic Q(S), function of random sample S;causes of nonresponse – impossibility for contact, refusal and unreturned questionnaires;application to compositional data - design-based theory, basic estimation approach used by data producers;Swiss Earnings Structure Survey (SESS) - in the SESS, D = 5 wage components recorded, being gross earnings
PB - Wiley
ER -