StatMatch R pakage
Marcello D’Orazio
Statistical Matching or Data Fusion
Overview
StatMatch is an R package for integrating two data sources referring to the same target population that share a number of common variables. Integration can be performed at:
- micro level — producing a synthetic (fused) file
- macro level — estimating correlation coefficients, regression coefficients, contingency tables, etc.
Some functions can also be used to impute missing values through hot deck imputation methods.
Installation
Install the stable version from CRAN:
install.packages("StatMatch")Main Features
- Non-parametric hot deck imputation: random, rank, and nearest-neighbour methods
- Predictive mean matching: a mixed parametric-nonparametric procedure
- Complex survey designs: methods for data from surveys with unequal weights
- Uncertainty analysis: tools to explore the uncertainty inherent in the statistical matching framework
Documentation & Tutorials
Additional material including vignettes, tutorials, and presentations is
available in the
Tutorials_Vignette_OtherDocs
folder.
References
- D’Orazio M., Di Zio M., Scanu M. (2006) Statistical Matching, Theory and Practice. Wiley, Chichester.
- D’Orazio M. (2015) Integration and imputation of survey data in R: the StatMatch package, Romanian Statistical Review, 2/2015, pp. 57–68.