Last modified: 2018-05-18
Abstract
Merging operations on two or more datasets has become an usual need for Statistical Institutes in the last few decades. Nowadays social sciences offer the opportunity of a new application of data integration techniques.
In this short paper we deal with the problem related to the estimation of the intergenerational earnings elasticity when proper datasets are not available. We compare the classical Two Samples Two Stages Least Squares with "Record Linkage'' and "Matching'' procedures.
References
1. Bjorklund A., Jantti M. (2011). Intergenerational income mobility and the role of family background.The Oxford Handbook of Economic Inequality.
2. Briscolini D., Di Consiglio L., Liseo B., Tancredi A., Tuoto T. (2017). New methods for smallarea estimation with linkage uncertainty. To appear on International Journal of ApproximateReasoning (IJAR).
3. Copas, J., Hilton, F. (1990). Record linkage: statistical models for matching computerrecords. Journal of the Royal Statistical Society, A, 153, pp. 287–320.
4. D’Orazio M., Di Zio M., Scanu M. (2004) Statistical matching: theory and practice, WileySeries in Survey Methodology, John Wiley & Sons, Ltd.
5. D’Orazio M. (2017) Statistical matching and imputation of survey data with Stat-Match for the R environment, R package vignette https:cran.r-project.orgwebpackages Stat-MatchvignettesStatistical Matching with StatMatch.pdf
6. Jerrim, J., Choi, A. and Rodriguez Simancas, R. (2016) Two-Sample Two-Stage Least Squares(TSTSLS) estimates of earnings mobility: how consistent are they? Survey Research Methods,Vol. 10, No. 2, pp. 85-102
7. Lahiri P., Larsen, M.D. (2005). Regression Analysis With Linked Data. Journal of the AmericanStatistical Association, 100, 222–230.
8. RodgersW.L. (1984) An evaluation of statistical matching, Journal of Business and EconomicStatistics 2, pp. 91102
9. Tancredi, A. and Liseo, B. (2015) Regression Analysis with linked data: Problems and possiblesolutions. Statistica, 75,1:19–35.