There are a few resources available for users who are familiar with Stata but want to move to R for one reason or another.

This particular page contains a useful reference of common Stata commands, along with comparable R commands for achieving the same task. Of course, the R commands will behave in slightly different ways than the Stata commands, so reading the documentation for each R command will be a prerequisite to using it. Note also that some of the commands are not included in the basic installation but will need to be installed before using them (typically through the install.packages() command).

Other useful resources include “The Stata Demo with R”, and the “Algorithm for selection of statistical methods”:

The Stata Demo with R is a replication of a Stata demo which users the “auto.dta” dataset to demonstrate the use of common statistical procedures. The demonstration session - as it looks when implemented in R - is available here:

• the Stata demo with R.

To find an R command using knowledge about the structure of a particular statistical problem, rather than the corresponding Stata command, you can use the algorithm for selection of statistical methods , which lists appropriate R commands by outcome variables and predictors. The algorithm is available here:

• Algorithm for selection of statistical methods

note: where the algorithm says glm(family=gaussian) it would probably be better just to use lm(). glm(family=gaussian) should work, but will be slightly less efficient etc.. (Perhaps this recommendation stems from a confusion between generalized linear models and general linear models?)