1Package: eco 2Version: 3.1-6 3Date: 2012-5-29 4Title: R Package for Ecological Inference in 2x2 Tables 5Author: Kosuke Imai <kimai@princeton.edu>, Ying Lu <yl46@nyu.edu>, 6 Aaron Strauss <abstraus@Princeton.EDU>. 7Maintainer: Ying Lu <yl46@nyu.edu> 8Depends: R (>= 2.0), MASS, utils 9Description: eco is a publicly available R package that implements the 10 Bayesian and likelihood methods proposed in Imai, Lu, and 11 Strauss (2008) for ecological inference in $2 \times 2$ tables 12 as well as the method of bounds introduced by Duncan and Davis 13 (1953). The package fits both parametric and nonparametric 14 models using either the Expectation-Maximization algorithms 15 (for likelihood models) or the Markov chain Monte Carlo 16 algorithms (for Bayesian models). For all models, the 17 individual-level data can be directly incorporated into the 18 estimation whenever such data are available. Along with 19 in-sample and out-of-sample predictions, the package also 20 provides a functionality which allows one to quantify the 21 effect of data aggregation on parameter estimation and 22 hypothesis testing under the parametric likelihood models. 23LazyLoad: yes 24LazyData: yes 25License: GPL (>= 2) 26URL: http://imai.princeton.edu/software/eco.html 27Packaged: 2012-05-29 15:44:23 UTC; kimai 28Repository: CRAN 29Date/Publication: 2012-06-12 06:36:54 30