Robert Wilkins
2009-Jan-09 00:56 UTC
[R] survey statistics, rate/proportions with standard errors
what does R have to compare with , say , proc surveymeans, estimate survey means/proportions with standard errors, using Taylor methods? [[alternative HTML version deleted]]
Stas Kolenikov
2009-Jan-09 03:37 UTC
[R] survey statistics, rate/proportions with standard errors
On 1/8/09, Robert Wilkins <iwritecode2 at gmail.com> wrote:> what does R have to compare with , say , proc surveymeans, estimate survey > means/proportions with standard errors, using Taylor methods?survey package by Thomas Lumley has pretty much everything you would need. Way more than SAS, at any rate. There are means, totals, ratios, as well as regression/GLM estimation procedures using Taylor series, jackknife, and the bootstrap. Try install.packages("survey") -- Stas Kolenikov, also found at http://stas.kolenikov.name Small print: I use this email account for mailing lists only.