Build Panel Data Sets from PSID Raw Data

Makes it easy to build panel data in wide format from Panel Survey of Income Dynamics ('PSID') delivered raw data. Downloads data directly from the PSID server using the 'SAScii' package. 'psidR' takes care of merging data from each wave onto a cross-period index file, so that individuals can be followed over time. The user must specify which years they are interested in, and the 'PSID' variable names (e.g. ER21003) for each year (they differ in each year). The package offers helper functions to retrieve variable names from different waves. There are different panel data designs and sample subsetting criteria implemented ("SRC", "SEO", "immigrant" and "latino" samples).


  • News in Version 1.6

  • Deprecated function get.variable.names

  • News in Version 1.5

  • Improve creation from text file example
  • Updates index to PSID version 2015
  • Automatically download wealth supplements if user requests years that contain a wealth supplement.

  • News in Version 1.4

  • Added ability to have NA values in the individual index
  • Don't download anything if files are present.
  • Added a real world example
  • Illustrates how variables can be constructed from a text file.

  • News in Version 1.3

  • News in Version 1.2

  • News in Version 1.1

  • updated index file to latest PSID delivery 2011
  • accomodates mutliple index files in datadir, with a warning
  • improved verbose mode
  • fixed a bug from using J() inside a data.table
  • added a not run example to help file

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


2.1 by Florian Oswald, 7 months ago

Browse source code at

Authors: Florian Oswald

Documentation:   PDF Manual  

Task views: Econometrics

GPL-3 license

Imports data.table, RCurl, foreign, SAScii, openxlsx, futile.logger

Suggests testthat

See at CRAN